CN115277567B - An intelligent reflective surface-assisted multi-MEC unloading method for Internet of Vehicles - Google Patents

Abstract

The invention discloses an intelligent reflective surface-assisted multi-MEC offloading method for the Internet of Vehicles, which includes: constructing a multi-mobile edge computing MEC collaborative Internet of Vehicles scenario that introduces intelligent reflective surface IRS; and based on the multi-MEC collaborative Internet of Vehicles scenario, constructing multi-MEC diversity offloading The model models the Internet of Vehicles user performance, IRS status and offloading selection as an optimization problem; by jointly optimizing local computing power, user transmission power, IRS phase and offloading RSU selection, the total number of processing bits is maximized; for the offloading The model, based on the idea of decoupling, calculates the optimal unloading solution. The technical solution of the present invention can solve the problem that Internet of Vehicles users often move to the edge of cells with poor signal quality, resulting in poor offloading links and reduced offloading efficiency.

Description

An intelligent reflective surface-assisted multi-MEC unloading method for Internet of Vehicles

Technical field

The invention relates to the technical field of Internet of Vehicles, and in particular to an intelligent reflective surface-assisted multi-mobile edge computing MEC offloading method for Internet of Vehicles.

Background technique

Internet of Vehicles (IoV) users will frequently experience moving to the edge of the cell due to their high mobility characteristics. There is a large signal quality gap between the center and the edge of a cell covered by traditional cellular coverage, which causes task offloading services for Internet of Vehicles users enabled with mobile edge computing (MEC) to experience fluctuations and even service interruptions. 6G hopes to achieve stable and seamless handover service quality. In order to solve this problem, existing research mainly focuses on improving the signal quality at the edge of the cell through physical layer technology, such as 6G cell-less technology.

In the IoV task offloading scenario where multiple MECs are introduced, most of the existing research on IoV mobility focuses on trajectory prediction or load balancing to avoid performing handover tasks at the edge of the cell or reduce the pressure on MECs. However, in the future, the business volume of the Internet of Vehicles will further increase, and various security and entertainment services will cover every minute and second of vehicle driving. Stopping task offloading when moving to the edge of the community will cause frequent fluctuations in service quality. Therefore, while improving signal quality, if a diversity offloading strategy is designed based on the characteristics of communication and computing resources at the edge of the cell, and roadside units (RSUs) with better channel conditions are selected for offloading, spectrum resources can be more fully utilized and further Improve service quality.

The existing technology mainly has the following shortcomings:

1) The existing technology ignores offloading services for users at the edge of the cell and cannot guarantee the stability and continuity of offloading services for high-mobility Internet of Vehicles users.

2) The existing technology improves the performance and reliability of task offloading through intelligent reflective surface (IRS)-assisted offloading, but does not propose a new multi-MEC collaborative offloading strategy based on the characteristics of the intelligent reflective surface, preventing IRS from exerting its best performance in task offloading. .

3) In the existing technology, the cells offloaded by Internet of Vehicles users are fixed and do not pay attention to the selection of roadside units (RSU) and MECs that assist in calculations, resulting in obvious fluctuations in the offloaded link as the user moves.

Contents of the invention

The present invention provides an intelligent reflective surface-assisted multi-MEC offloading method for Internet of Vehicles to solve the technical problem that Internet of Vehicles users often move to the edge of cells with poor signal quality, resulting in poor offloading links and reduced offloading efficiency.

In order to solve the above technical problems, the present invention provides the following technical solutions:

On the one hand, the present invention provides an intelligent reflective surface-assisted multi-MEC offloading method for the Internet of Vehicles. The intelligent reflective surface-assisted multi-MEC offloading method for the Internet of Vehicles includes:

Construct a multi-mobile edge computing MEC collaborative Internet of Vehicles scenario that introduces intelligent reflective surface IRS;

Based on the multi-MEC collaborative Internet of Vehicles scenario, a multi-MEC diversity offloading model is constructed to model the Internet of Vehicles user performance, IRS status and offloading selection as an optimization problem; by jointly optimizing local computing power, user transmission power, IRS phase and offloading Selection of RSU to maximize the total number of processed bits;

For the offloading model, based on the idea of decoupling, the optimal offloading solution is calculated.

Further, in the multi-MEC collaborative Internet of Vehicles scenario where IRS is introduced, there are multiple roadside unit RSUs equipped with MEC servers, and multiple smart vehicle users are randomly distributed within a certain range at the junction of the coverage areas of the two RSUs; user tasks It can be executed locally or offloaded to the MEC server of the RSU; an intelligent reflection plane is deployed at the midpoint of the connection between the two RSUs; the user's task offloading phase uses time division multiple access.

Further, the roadside unit RSU includes a first RSU and a second RSU.

Furthermore, the problem of maximizing the total number of processed bits is expressed as:

st

C2:0≤p _k,m ≤p _max

C3:0≤f _k,m ≤f _max

Among them, C _total is the total number of bits that the k-th user can process in the m-th time slot, C _total = C _local + C _offload , and C _local is the number of bits that the k-th user can process locally in the m-th time slot. The total number of bits, C _offload is the number of bits that the k-th user can complete processing through task offloading in the m-th time slot; p _k,m , f _k,m respectively represent the k-th user in the m-th time slot The transmit power and local _computing power in The upper limit of energy consumption in the gap, Select factors for uninstallation, /> When/> When , it means that the user offloads the task to the first RSU, when/> When , it means that the user offloads tasks to the second RSU; Θ _km is the IRS phase shift matrix; C1 represents the total energy limit of the user for local computing and transmitting offloading tasks in each time slot, τ is the length of each time slot, α is the computational complexity of the task, ξ is the user calculation energy consumption parameter; C2 and C3 are the definition domains of the transmit power p _k,m and the user's local computing capability f _k,m respectively, p _max represents the maximum value of the transmit power, f _max Indicates the maximum value of the user’s local computing capability; C4 indicates the offloading selection factor/> The value of is 0 or 1; C5 and C6 define the phase shift matrix and its constraints of the smart reflective surface, N is the total number of reflection units of the smart reflective surface, φ _k,m,n is the kth reflection unit in the mth The phase at the nth time slot of the user.

Furthermore, for the offloading model, the optimal offloading solution is calculated based on the idea of decoupling, including:

For the offloading model, based on the idea of decoupling, the IRS phase shift matrix is first optimized; then the phase shift matrix is substituted into the original problem to solve for the transmit power and user's local computing power; for the offloading selection factor, the problem is with/> Solve them separately, and select the offloading plan with a larger maximum number of processing bits after comparison.

Further, the optimized IRS phase shift matrix includes:

According to the triangle inequality, the optimal phase of each IRS unit is:

φ _k,m,n =arg(h _k,m )-arg(b _k,m [n])

Among them, φ _k,m,n is the phase of the k-th reflection unit at the n-th time slot of the m-th user, arg(*) refers to the phase factor, h _k,m is the current direct path channel status, b _{k ,m} [n] is the nth reflection unit of IRS, b _k,m = diag(G _k,m )h _r,k,m , where h _r,k,m is the channel status from kth user to IRS Information, G _k,m represents the channel status information from IRS to RSU.

Further, after substituting the phase shift matrix into the original problem, the transmit power and the user's local computing capability are solved, including:

After obtaining the optimization result of the phase shift matrix, suppose S=|h _k,m +G _k,m Θ _k,m h _r,k,m | ² and substitute S into the original problem:

st

C2:0≤p _k,m ≤p _max

C3:0≤f _k,m ≤f _max

Among them, B and σ ² are the channel bandwidth and noise power respectively;

After expressing the variable p _k,m by the variable f _k,m , the problem is equivalent to the following formula:

st

C3:0≤f _k,m ≤f _max

The analytical solution to the equivalent problem is obtained as:

in,

Further, after comparison, the offloading solution with a larger maximum number of processing bits is selected, including:

Calculate the total number of processing bits when offloading to the first RSU and the second RSU respectively, and compare the total number of processing bits when offloading to the first RSU and offloading to the second RSU. The larger value is the maximum required. The total number of processing bits, and the offloading decision factor is determined according to the offloading situation value, get the uninstallation plan.

In another aspect, the present invention also provides an electronic device, which includes a processor and a memory; wherein at least one instruction is stored in the memory, and the instruction is loaded and executed by the processor to implement the above method.

In another aspect, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored, and the instruction is loaded and executed by a processor to implement the above method.

The beneficial effects brought by the technical solutions provided by the present invention include at least:

1) The present invention introduces intelligent reflective surface IRS and proposes a new multi-MEC collaborative offloading strategy based on the characteristics of IRS. Compared with the existing technology, the IRS performance can be fully utilized, the Internet of Vehicles offloading strategy can be enriched, and the offloading efficiency of vehicle users can be improved.

2) The multi-MEC diversity offloading strategy adopted by the present invention can reduce the degradation of offloading performance caused by user channel state fluctuations. Compared with the existing technology, this technology selects the RSU for offloading according to the channel state, which can improve the quality and stability of the offloading link. Cell edge vehicle user offloading service quality.

Description of the drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without exerting creative efforts.

Figure 1 is a schematic diagram of a multi-MEC collaborative Internet of Vehicles scenario that introduces intelligent reflective surfaces;

Figure 2 is a schematic diagram of the change of the total number of user bits processed with respect to the number of IRS reflection units.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

First embodiment

In order to solve the problem that Internet of Vehicles users often move to the edge of cells with poor signal quality, resulting in poor offloading links and reduced offloading efficiency, this embodiment provides an intelligent reflector-assisted multi-MEC offloading method for Internet of Vehicles, which can Realized by electronic equipment, this method can, with the assistance of intelligent reflectors, enable cell edge users to select nearby roadside units for task offloading based on channel status to improve offloading link quality. The IoV user performance, IRS status, and offload RSU selection are modeled as an optimization problem, aiming to maximize the total number of user processing bits by jointly optimizing local computing power, user transmit power, IRS phase, and offload RSU selection. The execution process of this method includes the following steps:

S1, build a multi-mobile edge computing MEC collaborative Internet of Vehicles scenario that introduces intelligent reflective surface IRS;

Specifically, in this embodiment, a multi-MEC collaborative Internet of Vehicles scenario using intelligent reflective surfaces is shown in Figure 1 . It includes two roadside units equipped with MEC servers, and K smart vehicle users are randomly distributed within a certain range at the junction of the two RSU coverage areas. User tasks can be executed locally or offloaded to a roadside MEC server. However, no matter which RSU these users are offloaded to, they are located at the edge of the cell. The transmission distance is long and there are obstacles and other conditions, and the channel conditions are poor. Therefore, we deployed an intelligent reflection plane at the midpoint of the connection between the two RSUs. Users can improve the channel status through the superposition of direct paths and IRS paths. In order to avoid interference between users during the computation offloading process, the user's task offloading phase uses Time Division Multiple Access (TDMA), and the length of each time slot is τ. Each user is allocated M time slots to transmit or perform tasks locally. Among them, h _1,k,m and h _2,k,m are the direct path channel status information to RSU-1 and RSU-2 in the m-th time slot of the k-th user respectively. h _r,k,m is the channel state information from the k-th user to the IRS, and the channel state information from the IRS to RSU-1 and RSU-2 is G _1,k,m and G _2,k,m .

Let p _k,m and f _k,m respectively represent the transmit power and local computing capability of the k-th user in the m-th time slot. And E _max is the upper limit of the energy consumed by each user in a single time slot, then the energy constraint of each user in a single time slot is: Among them, α is the computational complexity of the task (cycles/bit), and ξ is the user calculation energy consumption parameter, which is related to the chip.

Then the total number of bits that the k-th user can process locally in the m-th time slot is:

By introducing intelligent reflective surfaces, the transmission channel from the user to the RSU superimposes the reflection path passing through the IRS on the basis of the direct path, improving the channel status. Therefore, the uplink transmission rates offloaded by user k to RSU-1 and RSU-2 in the m-th time slot are:

Among them, B and σ ² are the channel bandwidth and noise power respectively, Θ ₁ and Θ ₂ are the IRS phase shift matrices when offloading to RSU-1 and RSU-2 respectively. in

The smart reflective surface has a total of N reflection units, and φ _n is the phase of the nth reflection unit. By changing φ _n , the smart reflective surface can adjust the phase of the reflection path to achieve the effect of superposition or cancellation with the direct path.

Since the amount of result data is much smaller than the amount of task data, and the transmit power of the RSU is much greater than the transmit power of the vehicle, the downlink transmission delay will be much smaller than the uplink transmission delay. Therefore, the downlink transmission delay can be ignored.

S2, based on the multi-MEC collaborative vehicle networking scenario, construct a multi-MEC diversity offloading model, modeling the vehicle networking user performance, IRS status and offloading selection as an optimization problem; through joint optimization of local computing capabilities, user transmission power, IRS phase and the option to offload RSUs to maximize the total number of processed bits;

It should be noted that in order to improve the stability and service quality of task offloading and make full use of the characteristics of users being at the junction of two cells, we introduce an offloading selection factor Select the offloading target based on the channel status information. After comparing the user's current channel status to the two RSUs, select the better RSU for task offloading. When/> means that the user offloads the task to RSU-1, when/> means the user offloads the task to RSU-2. Then the total number of bits that can be transmitted to the offloading target RSU in one time slot is:

Among them, R _1,k,m and R _2,k,m are the uplink transmission rates of the k-th user offloaded to RSU-1 and RSU-2 in the m-th time slot respectively.

Since MEC has strong computing power and the amount of tasks uploaded in one time slot is small, we ignore the time it takes MEC to process tasks. And because the downlink transmission delay can be ignored, the number of bits processed through task offloading in one time slot is:

_Coffload ＝ _Ctransmit

On this basis, users can also jointly optimize transmit power and local computing power, allocate limited energy between task offloading and local computing, and achieve the goal of maximizing the total number of processed bits.

Therefore, the problem of maximizing the total number of processed bits can be expressed as:

st

C2:0≤p _k,m ≤p _max

C3:0≤f _k,m ≤f _max

Among them, C _total is the total number of bits that the k-th user can process in the m-th time slot, C _total =C _local +C _offload ; p _k,m , f _k,m respectively represent the k-th user in the m-th time slot. The transmit power and local computing power in each time slot; E _max is the upper limit of energy consumption by each user in a single time slot; the offloading selection factor is When/> means that the user offloads the task to RSU-1, when/> When represents the user offloading tasks to RSU-2; Θ _km is the IRS phase shift matrix; C1 represents the user's total energy limit for local computing and transmitting offloading tasks in each time slot, τ is the length of each time slot, α is the computational complexity of the task, ξ is the user calculation energy consumption parameter; C2 and C3 are the definition domains of the transmit power p _k,m and the user's local computing capability f _k,m respectively, p _max represents the maximum value of the transmit power, and f _max represents The maximum value of the user’s local computing capability; C4 represents the offload selection factor/> The value can be 0 or 1; C5 and C6 define the phase shift matrix and its constraints of the smart reflective surface, N is the total number of reflection units of the smart reflective surface, φ _k,m,n is the kth reflection unit in the mth The phase at the nth time slot of the user.

The change of the total number of user bits processed with respect to the number of IRS reflection units is shown in Figure 2.

S3. For the offloading model, based on the idea of decoupling, calculate the optimal offloading solution.

Specifically, in this embodiment, the above-mentioned S3 is specifically: for the above-mentioned offloading model, based on the idea of decoupling, a new alternating solution algorithm for timing decoupling in this scenario is proposed. In TDMA, each time slot is only allocated to one user, and the resources allocated to users are decoupled. Therefore, only a single time slot of the proposed problem is optimized separately, and then the solution to the original problem can be obtained by summing the terms. Since the proposed problem is characterized by time series decoupling and non-convexity, when solving the proposed optimization problem, the problem must first be decoupled according to time slots. After decoupling, each item of the problem is not jointly convex for each variable. We decompose each term of the original problem into two convex sub-problems and solve them alternately. First optimize the IRS phase shift matrix, then substitute the phase shift matrix into the original problem to solve for the transmit power and user's local computing power. For the unloading selection factor, we formulate the problem in with/> Solve them separately, and select the offloading plan with a larger maximum number of processing bits after comparison.

The specific optimization solution process is as follows:

S31, intelligent reflective surface phase shift matrix optimization.

Each reflection unit of the intelligent reflective surface can adjust the phase of the input IRS signal. When the phase of the IRS path signal and the direct path signal that finally reaches the RSU is the same, the amplitudes of the two signals are superimposed, and the channel state is optimal. Therefore, when offloading to RSU-1, the optimization problem of the phase shift matrix is as follows:

In order to solve the above problem, we let

|h _1,k,m +G _1,k,m Θ _1,k,m h _r,k,m |＝|h _1,k,m +θ _1,k,m b _1,k,m |

Among them, b _1,k,m =diag(G _1,k,m )h _r,k,m , |θ _k,m,n |=1. Then, according to the triangle inequality, we can get the upper bound of the problem

Among them, b _1,k,m [n] is the nth reflection unit of b _1,k,m . Then the upper bound of the phase factor is:

φ _k,m,n =arg(h _1,k,m )-arg(b _1,k,m [n])

Among them, arg(*) refers to the phase factor. Substitute φ _k,m,n into θ _1,k,m to obtain the optimal IRS channel state information θ _1,k,m b _1,k,m and |h _1,k +θ when offloaded to RSU-1 _1,k b _1,k | The phase shift matrix optimization when offloading to RSU-2 is the same, so it will not be described again here.

S32, user transmission power and computing power are optimized.

After obtaining the optimization result of the phase shift matrix, let S ₁ =|h _1,k,m +G _1,k,m Θ _1,k,m h _r,k,m | ² , S ₂ =|h _{2, k,m} +G _2,k,m Θ _2,k,m h _r,k,m | ² . Still taking the offloading to RSU-1 as an example, substituting S1 into the original objective function can be obtained:

st

C2:0≤p _k,m ≤p _max

C3:0≤f _k,m ≤f _max

By observing this problem, we can find that when the objective function reaches the maximum value, the constraint C1 will definitely take the equal sign, that is Therefore, according to/> After expressing the variable p _k,m by the variable f _k,m , the problem can be equivalent to the following formula:

st

C3:0≤f _k,m ≤f _max

Suppose the objective function of the above problem is F(f _k,m ), let Then the first-order and second-order derivatives of the objective function are respectively

Among them, F"(f _k,m )≤0 is always true, so the objective function F(f _k,m ) is a convex function. And because the st conditions C2 and C3 are obviously convex, therefore, the problem can be solved in F'(f _k The optimal solution is obtained when _,m )=0:

S33, based on the above, for each time slot of each user, according to the channel status from the user to RSU-1, first obtain the intelligent reflector phase shift matrix through phase shift matrix optimization, then substitute it into the original problem and then optimize by solving the convex optimization problem The user's transmit power and computing power are used to obtain the total number of processed bits in the time slot. In the same way, calculate the total number of processing bits when the task is offloaded to RSU-2, and compare the total number of processing bits when offloading to RSU-1 and offloading to RSU-2. The larger value is the maximum total number of bits sought. The number of processing bits can be processed, and the offloading decision factors can be determined based on the offloading situation. value, so use/> Dynamically select RSU to uninstall.

In summary, in order to solve the problem that Internet of Vehicles users often move to the edge of cells with poor signal quality, resulting in poor offloading links and reduced offloading efficiency, this embodiment proposes an intelligent reflective surface-assisted multi-MEC diversity offloading of Internet of Vehicles In this new offloading strategy, the transmission quality and task offloading rate are improved by introducing IRS, and by comparing the channel status of two adjacent cells, the RSU with better channel status is selected for offloading to reduce the direct path and the reflection path of IRS. The impact of two channel status fluctuations on offloading efficiency, thereby maximizing the number of processing bits for vehicle users at the edge of the cell. Moreover, active regulation of signals and offloading options can improve the quality and stability of offloaded links for IoV users at the edge of the community.

Second embodiment

This embodiment provides an electronic device, which includes a processor and a memory; wherein at least one instruction is stored in the memory, and the instruction is loaded and executed by the processor to implement the method of the first embodiment.

The electronic device may vary greatly due to different configurations or performance, and may include one or more processors (central processing units, CPU) and one or more memories, wherein at least one instruction is stored in the memory. The above instructions are loaded by the processor and execute the above method.

Third embodiment

This embodiment provides a computer-readable storage medium in which at least one instruction is stored, and the instruction is loaded and executed by a processor to implement the method of the first embodiment. The computer-readable storage medium may be ROM, random access memory, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. The instructions stored therein can be loaded by the processor in the terminal and execute the above method.

In addition, it should be noted that the present invention can be provided as a method, device or computer program product. Thus, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment that combines software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product embodied on one or more computer-usable storage media embodying computer-usable program code therein.

Embodiments of the invention are described with reference to flowcharts and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each process and/or block in the flowchart illustrations and/or block diagrams, and combinations of processes and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general-purpose computer, an embedded processor, or other programmable data processing terminal equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing terminal equipment produce a machine for A device that implements the functions specified in a process or processes in a flowchart and/or in a block or blocks in a block diagram.

These computer program instructions may also be stored in a computer-readable memory that causes a computer or other programmable data processing terminal equipment to operate in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction means, the The instruction means implements the functions specified in a process or processes of the flowchart and/or a block or blocks of the block diagram. These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, so that a series of operating steps are performed on the computer or other programmable terminal equipment to produce computer-implemented processing, thereby causing the computer or other programmable terminal equipment to perform a computer-implemented process. The instructions executed on provide steps for implementing the functions specified in a process or processes of the flow diagrams and/or a block or blocks of the block diagrams.

It should also be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations There is no such actual relationship or sequence between them. The terms "comprises," "comprises," or any other variation thereof are intended to cover a non-exclusive inclusion such that a process, method, article, or end device that includes a list of elements includes not only those elements, but also those not expressly listed Other elements, or elements inherent to the process, method, article or terminal equipment. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article or terminal device including the stated element.

Finally, it should be noted that the above descriptions are preferred embodiments of the present invention. It should be noted that although the preferred embodiments of the present invention have been described, for those skilled in the art, once the basic creative concept of the present invention is known, , without departing from the principles of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be regarded as the protection scope of the present invention. Therefore, it is intended that the appended claims be construed to include the preferred embodiments and all changes and modifications that fall within the scope of embodiments of the invention.

Claims (1)

1. An intelligent reflective surface-assisted multi-MEC offloading method for Internet of Vehicles, which is characterized by including: Construct a multi-mobile edge computing MEC collaborative Internet of Vehicles scenario that introduces intelligent reflective surface IRS; Based on the multi-MEC collaborative Internet of Vehicles scenario, a multi-MEC diversity offloading model is constructed to model the Internet of Vehicles user performance, IRS status and offloading selection as an optimization problem; by jointly optimizing local computing power, user transmission power, IRS phase and offloading Selection of RSU to maximize the total number of processed bits; For the unloading model, based on the idea of decoupling, the optimal unloading solution is calculated; In the multi-MEC collaborative vehicle networking scenario where IRS is introduced, there are multiple roadside unit RSUs equipped with MEC servers. Multiple smart vehicle users are randomly distributed within a certain range at the junction of the coverage areas of the two RSUs; the user's tasks can be performed locally Execute or offload to the MEC server of the RSU for execution; an intelligent reflection plane is deployed at the midpoint of the connection between the two RSUs; the user's task offloading phase uses time division multiple access; The roadside unit RSU includes a first RSU and a second RSU; The problem of maximizing the total number of bits processed is expressed as: Among them, C _total is the total number of bits that the k-th user can process in the m-th time slot, C _total = C _local + C _offload , and C _local is the number of bits that the k-th user can process locally in the m-th time slot. The total number of bits, C _offload is the number of bits that the k-th user can complete processing through task offloading in the m-th time slot; p _k,m , f _k,m respectively represent the k-th user in the m-th time slot The transmit power and local _computing power in The upper limit of energy consumption in the gap, Select factors for uninstallation,/> When/> When , it means that the user offloads the task to the first RSU. When When , it means that the user offloads the task to the second RSU; Θ _km is the IRS phase shift matrix; C1 represents the total energy limit of the user for local computing and transmitting offloading tasks in each time slot, τ is the length of each time slot, α is the computational complexity of the task, ξ is the user calculation energy consumption parameter; C2 and C3 are the definition domains of the transmit power p _k,m and the user's local computing capability f _k,m respectively, p _max represents the maximum value of the transmit power, f _max Indicates the maximum value of the user’s local computing capability; C4 indicates the offloading selection factor/> The value of is 0 or 1; C5 and C6 define the phase shift matrix and its constraints of the smart reflective surface, N is the total number of reflection units of the smart reflective surface, φ _k,m,n is the kth reflection unit in the mth The user’s phase at the nth time slot; For the offloading model, based on the idea of decoupling, the optimal offloading solution is calculated, including: For the offloading model, based on the idea of decoupling, the IRS phase shift matrix is first optimized; then the phase shift matrix is substituted into the original problem to solve for the transmit power and user's local computing power; for the offloading selection factor, the problem is with/> Solve them separately, and select the offloading plan with a larger maximum number of processing bits after comparison; The optimized IRS phase shift matrix includes: According to the triangle inequality, the optimal phase of each IRS unit is: φ _k,m,n =arg(h _k,m )-arg(b _k,m [n]) Among them, φ _k,m,n is the phase of the k-th reflection unit at the n-th time slot of the m-th user, arg(*) refers to the phase factor, h _k,m is the current direct path channel status, b _{k ,m} [n] is the nth reflection unit of IRS, b _k,m = diag(G _k,m )h _r,k,m , where h _r,k,m is the channel status from kth user to IRS Information, G _k,m represents the channel status information from IRS to RSU; After substituting the phase shift matrix into the original problem, the transmit power and the user's local computing power are solved, including: After obtaining the optimization result of the phase shift matrix, suppose S=|h _k,m +G _k,m Θ _k,m h _r,k,m | ² and substitute S into the original problem: Among them, B and σ ² are the channel bandwidth and noise power respectively; After expressing the variable p _k,m by the variable f _k,m , the problem is equivalent to the following formula: The analytical solution to the equivalent problem is obtained as: in, After comparison, select the offloading solution with a larger maximum number of processing bits, including: Calculate the total number of processing bits when offloading to the first RSU and the second RSU respectively, and compare the total number of processing bits when offloading to the first RSU and offloading to the second RSU. The larger value is the maximum required. The total number of processing bits, and the offloading decision factor is determined according to the offloading situation value, get the uninstallation plan.