Energy efficiency optimization method for D2D enhanced IRS assisted mobile edge computing system

Through D2D enhanced IRS-assisted mobile edge computing system, combined with D2D communication and IRS technology, the wireless communication signal blocking problem is solved, and communication reliability and data transmission efficiency are improved in complex environments, maximizing system energy efficiency.

CN119012275AInactive Publication Date: 2024-11-22ZHEJIANG SCI-TECH UNIV
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411497743.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-25
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In IoT applications, obstacles lead to blocking of wireless communication signals, resulting in unstable communication links, which are difficult for traditional technologies to effectively deal with. Especially in complex geographical environments, how to ensure the reliability and energy efficiency of communications has become a challenge.

Method used

Using D2D enhanced IRS-assisted mobile edge computing system, a mathematical model is established through the combination of D2D communication and IRS technology, and using the Lyapunov optimization and augmentation Lagrangian method to decompose optimization problems and optimize task offloading, transmission time, power and IRS beamforming maximizes system energy efficiency.

Benefits of technology

It effectively overcomes the problem of signal blocking, improves the reliability and coverage of wireless communication systems, accelerates data transmission, reduces task delay, maximizes system energy efficiency, and improves system performance and stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119012275A_ABST
    Figure CN119012275A_ABST
Patent Text Reader

Abstract

The invention discloses an energy efficiency optimization method of a D2D enhanced IRS assisted mobile edge computing system, relates to the technical field of wireless communication, and aims to solve the problem of signal obstruction caused by buildings and complex geographic landscapes in an Internet of Things environment, and effectively overcome signal obstruction and improve system performance by combining IRS and D2D technologies. According to task transmission and unloading processes of shielded users and mobile assistants, building a joint optimization problem by using a mathematical model, and evaluating the long-term performance of the system by taking energy efficiency as a key performance index; according to the method, Lyapunov optimization is adopted, an optimization problem is converted into a form easy to solve, a cooperative communication optimization penalty dual decomposition algorithm is designed, and a system energy efficiency optimization problem is decomposed into sub-problems to be solved. By adopting the method, the energy efficiency of the system is remarkably improved, the communication blocking problem of blocked wireless equipment in a severe environment can be solved, and the stability of a user queue is kept.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to an energy efficiency optimization method for a D2D enhanced IRS assisted mobile edge computing system. Background Art

[0002] With the rapid development of IoT technology, more and more smart devices are connected to the network, generating a large number of computing-intensive and delay-sensitive application requirements, such as smart factories, unmanned driving, virtual reality, and online games. These applications have put forward higher requirements on wireless communication networks, especially in wireless devices (WDs) with limited resources. How to efficiently complete computing tasks and data transmission has become an urgent problem to be solved.

[0003] However, in the actual IoT application environment, the presence of obstacles such as buildings and mountains will seriously block the wireless signal, causing unstable communication links and thus reducing the overall performance of the system. Especially in complex geographical environments with strong obstacles, the signal blocking problem is more obvious, and traditional wireless communication technologies are difficult to effectively cope with these challenges. Therefore, how to ensure the reliability and energy efficiency of communication in an obstacle environment has become a key issue in the design of IoT systems.

[0004] In order to solve the above challenges, intelligent reflecting surface (IRS) and device-to-device (D2D) cooperative communication technologies have gradually attracted attention. IRS is a technology that can manipulate the propagation path of wireless signals through programming. It can enhance signal coverage by reflecting wireless signals, especially when the signal is blocked by obstacles, which can effectively improve the communication quality. D2D communication technology allows direct communication between devices, reduces dependence on base stations, and can further improve the communication reliability and transmission efficiency of the system.

[0005] Therefore, there is an urgent need for an energy efficiency optimization algorithm based on a mobile edge computing system to simultaneously solve the problems of device communication signal obstruction and task queue congestion, and improve the transmission efficiency of the wireless communication network. Based on this, the present invention proposes a D2D enhanced IRS-assisted wireless power supply MEC system, which overcomes the signal blocking problem by combining IRS and D2D communication technologies, and improves the long-term performance and energy efficiency of the system. Summary of the invention

[0006] The purpose of the present invention is to provide an energy efficiency optimization method for a D2D enhanced IRS-assisted mobile edge computing system, which solves the problem of wireless communication signal obstruction caused by obstacles by combining D2D communication and IRS technology, and improves the communication reliability and data transmission efficiency of the system in complex environments.

[0007] To achieve the above object, the present invention provides a method for optimizing the energy efficiency of a D2D enhanced IRS-assisted mobile edge computing system, comprising the following steps: S1. For blocked mobile users, a D2D-enhanced IRS-assisted wireless power-supply mobile edge computing system model is established, so that the blocked users can pass tasks to the mobile assistant through D2D communication, and then offload them to the HAP; S2. According to the energy transmission and unloading process of the blocked user and the mobile assistant, a mathematical model is established, and a joint optimization problem is constructed with the system energy efficiency as the objective function; S3. Based on Lyapunov optimization, the joint optimization problem is transformed to decompose the long-term optimization problem into short-term optimization problems; S4. Using the augmented Lagrangian method, the transformed short-term optimization problem is transformed into a Lagrangian optimization problem, and then it is decomposed into sub-problems through the penalty dual decomposition method and solved in sequence.

[0008] Preferably, step S2 includes: S21. Establishing a task offloading model based on the blocked user and the mobile assistant; S22. Establishing a user computing model based on the task offloading model; S23, establishing a user energy model based on the collected and consumed energy; S24. Establish a queue model based on tasks or energy; S25. Define indicators of network stability and network energy efficiency, and use system energy efficiency as the objective function to establish a mathematical formula to describe the original optimization problem.

[0009] Preferably, in step S25, the original optimization problem is expressed as follows: ; ; In the formula, For system energy efficiency, is the number of single-antenna mobile users, is the number of blocked users, The length of a single time slot, superscript or subscript Indicates blocked users , superscript or subscript Represents mobile assistant , subscript Indicates user , superscript represents the maximum value of the corresponding label, , For users In time slot The transmission time and transmission power within , are the transmission time and transmission power of HAP, For users In time slot The local CPU frequency in For users The chip structure coefficient, For blocked users In time slot The uninstall decision , are the energy collected from the HAP by the blocked user and the mobile assistant, respectively. Energy harvesting efficiency for users, is the equivalent baseband channel from HAP to mobile assistant, and represents the phase shift vector of the blocked user’s uplink beam, and represents the mobile assistant uplink beam phase shift vector, , are the uplink and downlink IRS beam optimization variables respectively; and is the transmission time constraint that needs to be met, is the transmission power constraint of mobile users and HAP, is the local CPU frequency constraint of the mobile user. is the mobile user task queue constraint, , are the energy causality constraints of the blocked user and the mobile assistant, and is constrained by the blocked user task offloading policy, , are the phase shift constraints of IRS in downlink and uplink, respectively.

[0010] Preferably, step S3 includes transforming the original optimization problem based on Lyapunov optimization and processing the fractional form of the objective function to obtain the transformed optimization problem, which is expressed as follows: ; In the formula, is the weight parameter, For the system in time slot The total energy consumption, For users In time slot The task queue, For users In time slot The amount of tasks processed.

[0011] Preferably, it is characterized in that step S4 includes: S41, using the augmented Lagrangian method to process the discrete variables in the transformed optimization problem, and obtain the augmented Lagrangian optimization problem; S42, by using an effective block coordinate descent method, the augmented Lagrangian optimization problem is divided into an offloading strategy optimization subproblem, a transmission power and time optimization subproblem, a downlink beam optimization subproblem, an uplink beam optimization subproblem, and an auxiliary variable and local CPU frequency optimization subproblem, and the subproblems are solved in sequence; S43. An outer loop algorithm based on the penalty dual decomposition method updates the penalty factor and Lagrange multiplier to solve the optimal system energy efficiency.

[0012] Preferably, in step S41, the augmented Lagrangian optimization problem is expressed as follows: ; In the formula, is an auxiliary variable, and is the Lagrange multiplier, is the penalty factor.

[0013] Preferably, step S42 includes an inner loop algorithm based on effective block coordinate descent, using a convex optimization tool to solve individual sub-problems in sequence, as follows: The unloading strategy optimization sub-problem is expressed as: ; In the formula, is the system transmission bandwidth, is the equivalent baseband channel between the blocked user and the mobile assistant, , are the time slots of the blocked user and the mobile assistant respectively The task queue, , Blocked users transmission time and transmission power; Transmission power and time optimization subproblem, introducing auxiliary variables and Dealing with highly coupled variables, specifically expressed as: ; ; In the formula, , are the transmission time and transmission power of the mobile assistant respectively. For users In time slot The energy queue, For users Energy harvested from HAP; The downlink beam optimization subproblem is expressed as: ; in, ; ; ; In the formula, , is the auxiliary variable introduced; The uplink beam optimization subproblem is expressed as: ; In the formula, is the Gaussian white noise power at the HAP, is the Gaussian white noise power at the mobile assistant; The auxiliary variable and local CPU frequency optimization subproblem is expressed as: ; In the formula, is the introduced auxiliary variable.

[0014] Preferably, step S4 includes using the outer loop algorithm of the penalty dual decomposition method to update the Lagrange multiplier and the penalty factor, as follows: Penalty Factor pass , Update, where is a step size control parameter; The Lagrange multipliers are updated as follows: ; In the formula, is the PDD penalty parameter, is the number of outer loops; The violation constraint indicator for the optimization problem is: ; In the formula, Is a constraint violation indicator.

[0015] Therefore, the present invention adopts the above-mentioned D2D enhanced IRS assisted mobile edge computing system energy efficiency optimization method, which has the following technical effects: (1) By combining D2D communication and IRS technology, the signal blocking problem caused by buildings, geographical obstacles, etc. in the IoT environment is effectively overcome, and the communication reliability and coverage of the wireless communication system in complex environments are improved.

[0016] (2) Using smart reflective surfaces to enhance signal reflection and D2D technology for collaborative communication between devices can accelerate data transmission rates, reduce task transmission delays, and ensure that more tasks are processed in the same amount of time.

[0017] (3) The present invention maximizes the energy efficiency of the system by jointly optimizing task offloading, transmission time, power control, IRS beamforming and CPU frequency, thereby reducing energy consumption and improving the overall energy efficiency of the system.

[0018] (4) The present invention uses the penalty dual decomposition method to quickly and accurately solve complex joint optimization problems and efficiently obtain the optimal solution, thereby further improving the performance and stability of the system.

[0019] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic diagram of a D2D enhanced IRS assisted mobile edge computing system in an embodiment of an energy efficiency optimization method for a D2D enhanced IRS assisted mobile edge computing system; Figure 2 It is a schematic diagram of the PDD algorithm in an embodiment of the energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system. DETAILED DESCRIPTION

[0021] The present invention can be explained in more detail by the following examples. The purpose of disclosing the present invention is to protect all changes and improvements within the scope of the present invention. The present invention is not limited to the following examples.

[0022] like Figure 1 As shown, the energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system provided by the present invention includes the following steps: S1. Establish a D2D enhanced IRS-assisted wireless power supply mobile edge computing system model, including Single-antenna mobile users , , a hybrid access point (HAP) equipped with a radio frequency (RF) energy transmitter and MEC server, and a The IRS of the reflective elements can be express.

[0023] Mobile users The HAP can transmit information uplink to the HAP, and the HAP can transmit energy downlink to the mobile user. However, there is an obstacle between the mobile user and the HAP, which blocks the communication link between some mobile users and the HAP. Therefore, in this embodiment, the mobile users are divided into blocked users. , , and mobile assistant , Due to the existence of obstacles, the blocked users cannot communicate directly with the HAP and can only adjust the signal propagation path through the IRS or transmit it to the mobile assistant through D2D communication for processing.

[0024] Divide time into time slots, use Indicates that the length of each time slot is . Each time slot is subdivided into three stages: energy transmission stage, computation offloading stage and data backhaul stage. In this embodiment, in order to avoid interference in communications between mobile users, a time division multiple access (TMDA) communication method is adopted. In the energy transmission stage, the HAP broadcasts wireless energy signals to mobile users equipped with energy harvesting modules through a downlink. In the computation offloading stage, the process is divided into two parts: first, the blocked user can choose to offload the task to the AP through the IRS, or pass the task to the mobile assistant through D2D communication; secondly, the mobile assistant offloads the task to the HAP. In the data backhaul stage, due to the low latency characteristics of edge computing and the fact that the download speed is much higher than the task offloading speed, the duration of this stage is negligible.

[0025] S2. According to the mathematical model, a joint optimization problem is established, including the following steps: S21. Establish a task offloading model, as follows: In a time slot In the computation offloading phase, the blocked user needs to choose to reflect the offloading task to the HAP through IRS or transmit it to the mobile assistant through D2D communication. Indicates blocked users In time slot Uninstallation decision: When , indicating that in the time slot Blocked Users Mobile Assistant Perform D2D transmission. When, in the time slot Blocked Users No mobile assistant D2D transmission is performed. It should be noted that It does not indicate the time slot Blocked Users Reflected to HAP by IRS for unloading, only when When the blocked user Only then choose to uninstall through IRS reflection.

[0026] In order to avoid a sharp increase in the amount of tasks caused by the same mobile assistant receiving computing tasks from multiple blocked users in a certain time slot, it is necessary to ensure that the time slot Blocked users and mobile assistant The uniqueness of the D2D communication link between them is established by the following constraints: ; S22. Establish a user calculation model, as follows: (1) Local computing model Mobile users Equipped with not only an energy harvesting module but also an independent computing module, it is possible to choose to process some or all computing tasks locally. Since the computing and energy harvesting modules can work independently, mobile users can continue to perform local computing tasks throughout the time slot. The number of CPU cycles required to process each task, represents the local CPU frequency in the time slot, and the amount of tasks completed by the mobile user through local computing in the time slot It can be calculated as follows: ; (2) Blocked users uninstall the model when When the blocked user We choose to use IRS to offload the computational tasks to HAP through reflection. The equivalent baseband channels from HAP to IRS and from IRS to the blocked user are expressed as and , the blocked user is in the time slot With transmission power and transmission time Send information to the IRS, then the user is blocked In the The amount of tasks offloaded to HAP in time slots It is expressed as: ; In the formula, Indicates the system transmission bandwidth, is the Gaussian white noise power at the HAP, and represents the uplink beam phase shift vector of the IRS during the unloading of the blocked user. It should be noted that in the expression middle, Represents an imaginary unit rather than a mobile assistant .

[0027] when When the blocked user Choose to use D2D communication to offload the computing task to a mobile assistant. The equivalent baseband channel between the blocked user and the mobile assistant is Indicates that the blocked user In the The amount of tasks transmitted via D2D in a time slot for: ; In the formula, is the Gaussian white noise power at the mobile assistant.

[0028] In summary, blocked users In the The total number of tasks processed in a time slot for: ; (3) Mobile assistant uninstall model Unlike blocked users, mobile assistants Computational tasks can be offloaded through IRS reflection or directly to HAP.

[0029] The equivalent baseband channels from IRS to the blocked user and from HAP to the mobile assistant are expressed as and , move the assistant during uninstallation In time slot Transmit power and transmission time Send a message, then the mobile assistant In the The amount of tasks unloaded in a time slot It is expressed as: ; In the formula, and represents the uplink beam phase shift vector of the IRS during the mobile assistant unloading period. It should be noted that in the expression In each Previous represents the imaginary unit, and and In This means a mobile assistant .

[0030] However, since the blocked user may choose D2D communication to offload the computing task to the mobile assistant, the mobile assistant It is also necessary to consider the amount of tasks received from blocked users, which is expressed as follows: ; Therefore, the mobile assistant The total number of tasks processed in a time slot It can be expressed as: ; In summary, mobile users in time slots The amount of tasks processed for: ; S23. Establish a user energy model, as follows: During the energy transmission phase of a certain time slot, HAP transmits energy through the downlink at a certain transmission power. At the specified transfer time Therefore, HAP broadcasts wireless energy signals in this time slot. The energy consumption in the ; Assume that the efficiency of wireless energy harvesting of mobile users is , then the energy collected by the blocked user and the mobile assistant from the HAP is expressed as: and , that is, in a certain time slot user Energy harvested from HAP for: ; For mobile users, the purpose of collecting energy is to process tasks, and processing tasks will also consume energy. Generally speaking, its energy consumption is mainly generated by local computing and uplink offloading or D2D communication offloading. Represents mobile users The chip structure coefficient is obtained in the time slot user Energy consumption due to local computing for: ; Since the energy consumed by the blocked user for D2D communication offloading and IRS reflection offloading in the same time slot can be expressed by the same formula, the mobile user In time slot Energy consumed by transmission unloading It can be expressed as: ; Finally, for the convenience of calculation, the energy consumption of the IRS reflection unit and the HAP performing task calculations and transmitting the results are not considered.

[0031] Therefore, mobile users In time slot Energy consumption It can be expressed as: ; Throughout the time slot In the process, mobile users both collect and consume energy. If each mobile user has no energy storage at the beginning and the collected energy is greater than the consumed energy, then the mobile user In time slot The final energy obtained It is expressed as: ; In summary, the system is in the time slot Total energy consumption It can be expressed as: ; S24. Establish a queue model, as follows: use Represents mobile users The data task queue is used Indicates The size of the computational task that reaches the user at the end of the time slot. Without loss of generality, assume that and are independent and identically distributed in all time slots, and their average arrival rates are , , then the update rules of the task queue are as follows: ; Similarly, not only are there task queues, but energy can also be represented in the form of queues. Represents mobile users In time slot The energy queue of is updated as follows: ; Obviously, mobile users The energy consumed in a time slot cannot be greater than the time slot Battery remaining power and time slot before starting The sum of the energy collected by mobile users, so there is a constraint: ; S25. Establish a mathematical formula to describe the original optimization problem, as follows: The goal of this embodiment is to improve the long-term energy efficiency of the entire network while ensuring the stability of different types of task queues. To this end, in this embodiment, long-term network energy efficiency is used as the objective function, and indicators of network stability and network energy efficiency are defined. Among them, network stability reflects the ability of each task queue in the system to maintain low accumulation for a long time, avoiding excessive backlog of queues. Network energy efficiency measures the amount of effective tasks completed under unit energy consumption, thereby ensuring maximization of resource utilization and optimization of system performance.

[0032] Definition 1: A data queue with bounded average backlog time is stable, i.e. .

[0033] Definition 2: If each data queue in the network is stable, then the entire network queue is also stable.

[0034] Definition 3: The system energy efficiency is defined as the ratio between the long-term total energy consumption of the entire network and the total computing task size completed in a period of time, expressed as: ; In the formula, , Respectively , The limit of the time average expected value of The number of system users.

[0035] It should be noted that the definition of energy efficiency is very different from the common definition. In this embodiment, the concept of time is taken into consideration and it is defined from a long-term perspective. Therefore, it can describe both energy efficiency and average delay at the same time.

[0036] Taking system energy efficiency as the objective function, the original optimization problem is expressed as:

[0037] In the formula, , and represent the IRS beam optimization variables for uplink and downlink, respectively, and is the transmission time constraint that needs to be met, represents the transmission power constraints of mobile users and HAPs, Indicates the local CPU frequency constraint of the mobile user. represents the mobile user task queue constraint. In addition, and represent the energy causality constraints of the blocked user and the mobile assistant, respectively. and represents the task offloading policy constraint of the blocked user, and, and are the phase shift constraints of IRS in downlink and uplink, respectively.

[0038] Since this optimization problem spans multiple time slots, it is a multi-time slot stochastic optimization problem involving joint resource allocation of multiple variables. It requires the coordinated scheduling of multiple users and the design of multiple variable optimizations, including task offloading strategies, transmission power allocation, transmission time allocation, IRS phase shift vectors, and joint optimization of local CPU frequency. These variables are highly coupled to each other and contain both discrete and continuous variables, which increases the complexity of the problem. Therefore, compared with traditional mixed integer nonlinear programming problems, the difficulty of solving this problem is significantly increased, and it is a typical NP-hard problem.

[0039] S3. Since the objective function is in a non-convex fractional form, this usually leads to difficulty in solving. Based on this, in this embodiment, the objective function is first transformed using the properties of fractional optimization to simplify the solution process. Specifically, by performing an equivalent transformation on the objective function in fractional form, the original non-convex problem is transformed into a more easily tractable form, as shown below: Theorem 1: System Energy Efficiency is optimal if and only if .

[0040] According to Theorem 1, the original optimization problem is equivalently converted to:

[0041] in, represents the optimal solution of the optimization problem. Although a relatively simplified optimization problem is obtained through equivalent transformation, the problem still has certain difficulties, mainly because The value is still unknown, which makes the problem still have a high complexity of solution, especially when dealing with joint optimization involving multiple variables, so it needs to be further transformed.

[0042] Since before the time slot starts, the total energy consumption of the entire network and the total amount of computing tasks completed Easy to obtain, therefore, can be obtained in the time slot Before starting, the system efficiency is: ; In the formula, For the system in time slot The total energy consumption before the start. For the system in time slot The total amount of tasks to be processed before starting. Assume , then available The system is in the time slot Total energy efficiency before starting, replacement , the optimization problem is further transformed into:

[0043] Although the fractional form of the objective function has been processed, the processed optimization problem is still a long-term optimization problem. To solve this problem, this embodiment introduces the Lyapunov optimization technique to transform the long-term random optimization problem into a series of short-term deterministic problems that can be solved hour by hour. Through this transformation, it is possible to optimize independently in each time slot, thereby greatly simplifying the problem-solving process, while ensuring the long-term performance of the system and ensuring the stability of the system.

[0044] Among them, the Lyapunov function is defined as: ; The Lyapunov drift plus penalty function is defined as: ; in, is an adjustable weight parameter, which is the relative importance of system energy efficiency and latency. To ensure the stability of the network, we will Minimize, which also means that the system energy efficiency is minimized. Then, we derive from the following theorem The upper bound of .

[0045] The upper bound of this problem is derived as follows: ; in, ; In the formula, and They are and The upper bound of .

[0046] According to the principle of minimizing expected opportunities, the optimization problem in each time slot is further simplified by minimizing the upper bound on the right side of the inequality. In this way, the optimal solution for system energy efficiency can be solved in each time slot, decomposing the complex long-term optimization problem into multiple short-term problems. Finally, the optimization problem can be transformed into the following form: ; S4, such as Figure 2 As shown, based on the penalized dual decomposition (PDD) method, the transformed optimization problem is solved, including the following steps: S41. Using the augmented Lagrangian method, the discrete variables in the transformed optimization problem are processed to obtain the augmented Lagrangian optimization problem, which is as follows: It is not difficult to find that the optimization problem after Lyapunov optimization transformation has discrete variables , making the problem difficult to handle. In this embodiment, an auxiliary variable is introduced , constrain The equivalent transformation is: ; Then, using the augmented Lagrangian method, the transformed equation constraints are penalized into the objective function, and the augmented Lagrangian optimization problem is obtained, which is expressed as: ; In the formula, and is the Lagrange multiplier, is the penalty factor.

[0047] In the PDD framework, the Lagrangian multipliers and penalty factors are kept fixed in the inner loop and are not updated; their update operations are performed in the outer loop of PDD. When the update in the outer loop continues until convergence, the augmented Lagrangian problem can be regarded as equivalent to the original optimization problem. This step-by-step approximation method ensures that when the convergence condition is reached, the optimal solution can well approximate the optimal solution of the original problem, thereby achieving an effective solution to the problem.

[0048] S42. Through the effective block coordinate descent method, in the inner loop of the PDD method, the augmented Lagrangian optimization problem is divided into sub-problems and solved in sequence, as follows: (1) Under the condition that other variables are fixed, solve the unloading strategy optimization sub-problem. The specific expression is: ; The variables to be optimized are the offloading decision variables for coordinating D2D transmission with multiple users or transmitting through IRS. This optimization subproblem is a convex optimization problem with inequality constraints, which can be solved using various convex optimization tools and classical algorithms, such as CVX and CPLEX toolboxes.

[0049] (2) Solve the transmission power and time optimization sub-problem. The specific expressions are as follows: ; ; Since the transmission power and transmission time variables are highly coupled, in order to deal with the difficulties caused by the highly coupled variables, in this embodiment, a set of auxiliary variables is introduced and , the optimization subproblem is reformulated as: ; ; Among them, based on the properties of the perspective function, the function after perspective is still a concave function. yes The concave function of Too and It is a concave function, so it can be solved using convex optimization tools such as CVX and CPLEX.

[0050] (3) Solve the downlink beam optimization subproblem. The specific expression is as follows: ; make , , , , , , , and needs to meet and , we can get: ; In the formula, , is the introduced auxiliary variable.

[0051] This subproblem can be reformulated as: ; Since this subproblem has a rank-one constraint, the SDR method is used to relax it and transform it into a convex problem, and then a standard convex optimization solver (such as CVX) is used to optimize the solution.

[0052] (4) Solve the uplink beam optimization sub-problem, which is specifically expressed as follows: ; ; It can be found that when the IRS reflected signal is aligned with the IRS reflected signal on the AP user link, it is the optimal vector of the IRS uplink phase shift, so we get: ; (5) Solve the auxiliary variable and local CPU frequency optimization sub-problem, which is specifically expressed as follows: ; Since the auxiliary variable and mobile user local CPU frequency They do not interfere with each other and are independent of each other, so they can be solved separately.

[0053] Firstly, for For example, it is a typical unconstrained convex quadratic programming problem, and the closed-form solution can be directly obtained as shown below: ; and The optimal solution of is located at its boundary or stationary point, and its closed-form solution can be expressed as: ; S43. An outer loop algorithm based on the penalty dual decomposition method updates the penalty factor and Lagrange multiplier to solve the optimal system energy efficiency.

[0054] Among them, the penalty factor pass , Update, where is a step size control parameter.

[0055] The Lagrange multipliers are updated as follows: ; In the formula, is the PDD penalty parameter, is the number of outer loops.

[0056] The violation constraint indicator for the optimization problem is: ; In the formula, Is a constraint violation indicator.

[0057] Therefore, the present invention adopts the above-mentioned D2D enhanced IRS assisted mobile edge computing system energy efficiency optimization method to solve the data transmission problem of blocked users, solve the problem quickly and accurately, and achieve the improvement of system energy efficiency.

[0058] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. The energy efficiency optimization method of D2D enhanced IRS assisted mobile edge computing system is characterized by: The following steps are involved: S1. For blocked mobile users, a D2D-enhanced IRS-assisted wireless power-supply mobile edge computing system model is established, so that the blocked users can pass tasks to the mobile assistant through D2D communication, and then offload them to the HAP; S2. According to the energy transmission and unloading process of the blocked user and the mobile assistant, a mathematical model is established, and a joint optimization problem is constructed with the system energy efficiency as the objective function; S3. Based on Lyapunov optimization, the joint optimization problem is transformed to decompose the long-term optimization problem into short-term optimization problems; S4. Using the augmented Lagrangian method, the transformed short-term optimization problem is transformed into a Lagrangian optimization problem, and then it is decomposed into sub-problems through the penalty dual decomposition method and solved in sequence.

2. The energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system according to claim 1 is characterized in that: Step S2 includes: S21. Establishing a task offloading model based on the blocked user and the mobile assistant; S22. Establishing a user computing model based on the task offloading model; S23, establishing a user energy model based on the collected and consumed energy; S24. Establish a queue model based on tasks or energy; S25. Define indicators of network stability and network energy efficiency, and use system energy efficiency as the objective function to establish a mathematical formula to describe the original optimization problem.

3. The energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system according to claim 2 is characterized in that: In step S25, the original optimization problem is expressed as follows: ; In the formula, For system energy efficiency, is the number of single-antenna mobile users, is the number of blocked users, The length of a single time slot, superscript or subscript Indicates blocked users , superscript or subscript Represents mobile assistant , subscript Indicates user , superscript represents the maximum value of the corresponding label, , For users In time slot The transmission time and transmission power within , are the transmission time and transmission power of HAP, For users In time slot The local CPU frequency in For users The chip structure coefficient, For blocked users In time slot The uninstall decision , are the energy collected from the HAP by the blocked user and the mobile assistant, respectively. Energy harvesting efficiency for users, is the equivalent baseband channel from HAP to mobile assistant, and represents the phase shift vector of the blocked user’s uplink beam, and represents the mobile assistant uplink beam phase shift vector; and is the transmission time constraint that needs to be met, is the transmission power constraint of mobile users and HAP, is the local CPU frequency constraint of the mobile user. is the mobile user task queue constraint, and are the energy causality constraints of the blocked user and the mobile assistant, and is constrained by the blocked user task offloading policy, and are the phase shift constraints of IRS in downlink and uplink, respectively.

4. The energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system according to claim 1, characterized in that: In step S3, the original optimization problem is transformed based on Lyapunov optimization, and the fractional form of the objective function is processed to obtain the converted optimization problem, which is expressed as follows: ; In the formula, is the weight parameter, For the system in time slot The total energy consumption, For users In time slot The task queue, For users In time slot The amount of tasks processed.

5. The energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system according to claim 1, characterized in that: Step S4 includes: S41, using the augmented Lagrangian method to process the discrete variables in the transformed optimization problem, and obtain the augmented Lagrangian optimization problem; S42, by using an effective block coordinate descent method, the augmented Lagrangian optimization problem is divided into an offloading strategy optimization subproblem, a transmission power and time optimization subproblem, a downlink beam optimization subproblem, an uplink beam optimization subproblem, and an auxiliary variable and local CPU frequency optimization subproblem, and the subproblems are solved in sequence; S43. An outer loop algorithm based on the penalty dual decomposition method updates the penalty factor and Lagrange multiplier to solve the optimal system energy efficiency.

6. The energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system according to claim 5 is characterized in that: In step S41, the augmented Lagrangian optimization problem is expressed as follows: ; In the formula, is an auxiliary variable, and is the Lagrange multiplier, is the penalty factor.

7. The energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system according to claim 6, characterized in that: In step S42, an inner loop algorithm based on effective block coordinate descent is included, and a convex optimization tool is used to solve the individual sub-problems in sequence, as follows: The unloading strategy optimization sub-problem is expressed as: ; In the formula, is the system transmission bandwidth, is the equivalent baseband channel between the blocked user and the mobile assistant, , are the time slots of the blocked user and the mobile assistant respectively The task queue, , Blocked users transmission time and transmission power; Transmission power and time optimization subproblem, introducing auxiliary variables and Dealing with highly coupled variables, specifically expressed as: ; ; In the formula, , are the transmission time and transmission power of the mobile assistant respectively. For users In time slot The energy queue, For users Energy harvested from HAP; The downlink beam optimization subproblem is expressed as: ; in, ; ; ; In the formula, , is the auxiliary variable introduced; The uplink beam optimization subproblem is expressed as: ; In the formula, is the Gaussian white noise power at the HAP, is the Gaussian white noise power at the mobile assistant; The auxiliary variable and local CPU frequency optimization subproblem is expressed as: ; In the formula, is the introduced auxiliary variable.

8. The energy efficiency optimization method of the D2D enhanced IRS assisted mobile edge computing system according to claim 7, characterized in that: In step S4, the outer loop algorithm of the penalty dual decomposition method is used to update the Lagrange multiplier and the penalty factor, as follows: Penalty Factor pass , Update, where is a step size control parameter; The Lagrange multipliers are updated as follows: ; In the formula, is the PDD penalty parameter, is the number of outer loops; The violation constraint indicator for the optimization problem is: ; In the formula, Is a constraint violation indicator.

Citation Information

Patent Citations

  • Energy efficiency optimization method of D2DMEC system based on IRS assistance

    CN115802419A

  • Maximum sum rate resource optimization method for IRS-assisted cognitive SWIPT system

    CN116321186A

  • High-reliability task unloading and resource management algorithm under 5G-WiFi heterogeneous network

    CN118250748A