IRS-aided radio frequency charging UD-MEC system task offloading method

Abstract

The present application relates to the field of wireless communication network and super dense edge computing technology, specifically to an IRS assisted radio frequency charging UD-MEC system task offloading method, comprising: S1: dividing mobile users and base stations into several clusters; S2: calculating the time and energy required for processing tasks; S3: based on IRS assistance, minimizing the total energy consumption of mobile users in several task scenarios, the task scenarios including a regular scenario, a scenario that can only be divided into several fixed threads or methods, and a complex scenario that can only be divided into several fixed threads or methods, and some sub-tasks can only be processed locally; S4: jointly considering the service cache problem of the base station, minimizing the total energy consumption of the mobile user; S5: selectively using millimeter wave, terahertz frequency band, deploying UAV as a movable base station, and minimizing the energy consumption of the UAV. The technical scheme can effectively reduce the energy consumption of mobile user task offloading.

Description

Technical Field

[0001] This invention relates to the fields of wireless communication networks and ultra-dense edge computing technology, specifically to a task offloading method for an IRS-assisted radio frequency charging UD-MEC system. Background Technology

[0002] The 6G era is an era of ubiquitous connectivity, with global mobile communication data volume expected to reach 1000 times that of the 5G era. As mobile users become increasingly intelligent, computationally intensive applications, such as virtual reality and remote e-health, are placing ever greater demands on computing power. Ultra-dense networks and multi-access edge computing (MEC) are considered two effective technologies capable of addressing the massive data transmission, microsecond-level communication latency, ubiquitous connectivity, and rapidly growing computing demands of the 6G era. The combination of these two technologies has given rise to ultra-dense multi-access edge computing (UD-MEC).

[0003] Ultra-dense networks typically deploy various types of low-power small cell base stations (SBSs), such as picocells, femtocells, WiFi access points (WiFi APs), and even unmanned aerial vehicles (UAVs), in extremely dense deployments in hotspot areas. This heterogeneous ultra-dense networking can effectively eliminate signal blind spots, reduce user service latency, and improve spectrum utilization and system capacity. Meanwhile, to meet the needs of massive numbers of users for sub-second (<1ms) or even microsecond-level communication latency and ubiquitous connectivity, multi-access edge computing allows mobile users to offload computing tasks to one or more MEC servers at different wireless access points using various wireless access technologies.

[0004] Ultra-dense networks (UD-MEC) and multi-access edge computing complement each other. By combining the two and deploying MEC servers at base stations, an UD-MEC system can be formed. On the one hand, it can provide mobile users with a large number of wireless access points to meet the requirements of the Internet of Things in the 6G era; on the other hand, it can provide them with a large amount of computing resources. Although there are already some achievements in UD-MEC, due to factors such as wireless channel fading, limited communication resources, and obstacles, the communication latency of mobile users in the UD-MEC system is still difficult to meet the microsecond-level latency requirements of the 6G era. Therefore, based on key 6G technologies such as millimeter wave and terahertz communication, how to significantly improve the task offloading rate of massive users in the UD-MEC system and design a reasonable task offloading strategy are urgent problems to be solved.

[0005] Intelligent Reflecting Surfaces (IRS) are considered a revolutionary technology that can effectively solve the aforementioned problems. An IRS consists of a large number of intelligent reflecting units, each of which independently reflects the incident signal by controlling its own phase and / or amplitude, and intelligently configures the wireless communication environment to achieve purposes such as improving information transmission rates or enabling signals to bypass obstacles. IRSs have low deployment costs and can operate in full-duplex mode with microwatts of power consumption, far lower than that of relay nodes in full-duplex communication. Therefore, for UD-MEC systems, IRS reflectors (usually placed on building surfaces) can be deployed near the base station and / or mobile user terminal to improve UD-MEC system performance. However, how to rationally deploy IRS reflectors in UD-MEC systems and dynamically and intelligently select parameters such as the phase and / or amplitude of the IRS to improve the task offloading rate and RF energy harvesting efficiency of the UD-MEC system are still problems that urgently need to be solved. Summary of the Invention

[0006] The purpose of this invention is to propose a task offloading method for an IRS-assisted radio frequency charging UD-MEC system. This technical solution can effectively reduce the energy consumption of mobile user task offloading.

[0007] To achieve the above objectives, the present invention provides a basic solution: a task offloading method for an IRS-assisted radio frequency charging UD-MEC system, comprising:

[0008] S1: Divide mobile users and base stations into several clusters;

[0009] S2: Calculates the time and energy required to process the task;

[0010] S3: Based on IRS assistance, minimize the total energy consumption of mobile users in several task scenarios. The task scenarios include regular scenarios, scenarios that can only be divided into several fixed threads or methods, and complex scenarios that can not only be divided into several fixed threads or methods, but also some sub-tasks can only be processed locally.

[0011] S4: Jointly consider the service caching problem of base stations to minimize the total energy consumption of mobile users;

[0012] S5: Selectively use millimeter wave and terahertz frequency bands to deploy UAVs as mobile base stations and minimize UAV energy consumption.

[0013] Beneficial effects of the basic scheme: This scheme utilizes the intelligent reflection of incident signals by the IRS and the dynamic adjustment of the wireless communication environment, which greatly improves the task offloading rate of mobile users; clustering is performed first in S1, which can reduce interference between users and the complexity of subsequent solutions; S3 and S4 minimize several task scenarios and jointly consider the base station's caching problem, which minimizes the energy consumption of mobile users, reduces the system's operating power consumption, thereby extending the device's usage time and battery life, and can maintain the device's uninterrupted operation.

[0014] The use of millimeter wave and terahertz frequency bands in S5 can effectively solve the co-channel interference problem caused by too many small base stations in the UD-MEC system. However, since the power consumption of terahertz and millimeter wave frequency bands increases with the frequency, they are only used selectively in some coverage areas and some time periods to reduce system power consumption while ensuring the smooth completion of task offloading.

[0015] To reduce the deployment of small base stations and address the issues of millimeter wave and terahertz bands being highly susceptible to obstruction and signal attenuation due to their extremely high frequencies and short wavelengths, thus limiting communication range, this solution also deploys UAVs as mobile base stations. This ensures efficient task offloading, while minimizing UAV power consumption to extend UAV usage time and ensure system stability, thereby effectively improving the user's task offloading rate.

[0016] As a preferred embodiment, in S1, mobile users in each cluster can communicate with all small base stations within that cluster, and each cluster selects one small base station as the cluster head.

[0017] Mobile users in each cluster communicate with small base stations within the cluster, reducing co-channel interference caused by direct communication outside the cluster. By selecting a cluster head to manage and allocate tasks and manage nodes within the cluster, and exchanging and updating information with the external network, smooth communication is ensured.

[0018] As a preferred option, several small base stations are selected in each cluster as the energy source for mobile users or radio frequency charging small base stations within the cluster.

[0019] By selecting and setting specific energy sources, network coverage can be improved, communication efficiency can be enhanced, and network construction costs can be reduced.

[0020] S2 includes:

[0021] S21; Calculate the time and energy consumption of mobile users, assuming mobile users... Information transmission time is ,get Total time required to process the task and total energy They are respectively:

[0022]

[0023]

[0024] in Indicates the clusters of partitions, They are respectively The time and energy required for local task processing for to base station Energy consumption of unloading tasks and base stations The time required to process the task;

[0025] S22: Calculate the energy required for the base station to process the task. For mobile users The energy consumed in processing the task is:

[0026] ,

[0027] in for The rate at which tasks are unloaded, due to With base station The channel gain between the two is related to the phase and amplitude of the IRS. It is closely related to the phase and amplitude of the IRS, and It is a non-convex, non-concave function of the phase and amplitude of the IRS.

[0028] The calculation of mobile user time and energy under normal conditions, as well as the energy of base station processing tasks, provides parameters for subsequent energy consumption minimization.

[0029] As a preferred embodiment, S3 includes:

[0030] S31: In a typical scenario, to minimize the total energy consumption of mobile users, construct a task offloading model for the following problem P1:

[0031]

[0032] in and Phase and amplitude constraints are given for IRS in both energy acquisition and information transmission, where , The first The phase of an IRS during auxiliary energy acquisition and information transmission. and The corresponding amplitude; and users respectively and base stations Energy constraints; For users Task execution time constraints and task quantity constraints; For users Power constraints; For base stations CPU constraints for task execution;

[0033] S32: In mobile users In scenarios where a task can only be divided into a fixed number of threads or methods, each thread or method can be considered a subtask. express Should subtasks be included? Unload to base station ,in express Subtasks Unload to base station ,otherwise Add the following constraints to the task unloading model for problem P1:

[0034]

[0035] in Each subtask The amount of data is And each subtask It can only be processed locally or offloaded to a small base station for processing, among which To optimize variables.

[0036] S31 and S32 solve the problem of minimizing energy consumption in conventional scenarios and in non-complex scenarios that can only be divided into a few fixed threads or methods.

[0037] As a preferred embodiment, S3 further includes:

[0038] S33: In complex scenarios, mobile users The task can not only be divided into a fixed number of threads or methods, but some subtasks can only be processed locally. During specific processing, a call graph is used. To model the relationships between subtasks, the graph includes a set of vertices consisting of all subtasks, which are used to represent the dependencies between subtasks. Dependencies include local processing relationships, parallel processing relationships, and sequential processing relationships.

[0039] As a preferred embodiment, S4 includes:

[0040] S41: Let the set of service programs be... Due to base stations Since the MEC server cache space is limited, the following constraint is added to the task offloading model when calculating the minimum energy consumption:

[0041] ,

[0042] in , At that time, base station The MEC server caches the service program at the specified location, and vice versa;

[0043] S42: Minimize the total energy consumption of mobile users in scenarios where users need to upload service programs to the MEC server, or where small base stations need to transmit service programs wirelessly.

[0044] With the assistance of IRS, the latency of service program download can be greatly reduced. At the same time, since the cache space of MEC server at the base station is limited, IRS technology will affect the service cache of MEC server. S41 takes into account the MEC server cache space to calculate the minimum energy consumption, and then minimizes the energy consumption of mobile users in special scenarios through S42.

[0045] As a preferred embodiment, S42 includes:

[0046] S42-1: Selective caching service program based on task computation requirements;

[0047] S42-2: To minimize the total energy consumption of mobile users, construct a task offloading model for the following problem P2:

[0048]

[0049] Among them, the task quantity constraints in problem P1 Modifications were made to add constraints on the uploading of service program data.

[0050] Since the MEC server at the base station has limited cache space, S42-1 filters service programs with higher cache requirements based on computational needs, eliminating some less demanding cache tasks. This directly reduces cache pressure and saves subsequent computational resources. S42-2 addresses the issue of minimizing energy consumption when jointly considering service caching by modifying the model constraints for offloading subtasks to the base station.

[0051] As a preferred embodiment, S5 includes:

[0052] S51: Deploys spatially mobile base stations and uses UAVs as MEC servers to dynamically execute tasks for mobile users;

[0053] S52: Divide the terahertz frequency band used into Each sub-band, time divided into In each time slot, the user selects a different terahertz sub-band, and the ground user, with the assistance of the IRS, offloads the task to the UAV for processing via the terahertz band;

[0054] S53: To minimize the energy consumption of the UAV, construct a model for problem P3 as follows:

[0055]

[0056] in, For each time slot Time constraints and Position constraints for UAVs Unload constraints for the task. For power and energy constraints for mobile users, Assign constraints to sub-bands. Assign constraints to the CPU of the UAV.

[0057] S51 By deploying mobile base stations, the deployment density of small base stations can be minimized; S52 can improve signal reliability and transmission efficiency while ensuring signal clarity; S53 Minimizing UAV energy consumption can extend UAV usage time and ensure system communication quality. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of an IRS-assisted radio frequency charging UD-MEC system;

[0059] Figure 2 This is a logic diagram of a task offloading method for an IRS-assisted RF charging UD-MEC system;

[0060] Figure 3This is a diagram illustrating a subtask call example;

[0061] Figure 4 This is a diagram illustrating task offloading and service caching in edge computing;

[0062] Figure 5 This is a schematic diagram of a terahertz UD-MEC system assisted by IRS and UAV. Detailed Implementation

[0063] The technical solution of this application will be further described in detail below through specific embodiments:

[0064] like Figure 1 The illustrated IRS-assisted RF charging UD-MEC system includes a macro base station. , Small base stations , mobile users Each base station is equipped with an MEC server, and the set of base stations is denoted as . The mobile user set is Each mobile user All are rechargeable nodes. In each time slot... (Time slot length is) ), First, radio frequency energy (duration is) is obtained from the radio frequency signal of the energy source. The system then utilizes the acquired energy to perform computational tasks. The set of radio frequency charging small base stations in the system is... Furthermore, each small base station is equipped with multiple antennas, thus Small and medium-sized base stations utilize antenna switching technology to select some antennas to transmit information and others to receive energy, based on the current antenna efficiency, in order to simultaneously acquire energy and receive information.

[0065] Due to the ultra-dense deployment of heterogeneous small base stations, mobile users can access multiple base stations to offload tasks, such as... Figure 2 As shown, the specific steps for task uninstallation are as follows:

[0066] S1: Divide mobile users and base stations into several clusters according to certain criteria, such as geographical location, to reduce interference between users and solution complexity. Let the set of all clusters be denoted as . In a cluster, mobile users can communicate with all small base stations within that cluster, and each cluster selects one small base station as the cluster head. Assume... Belongs to cluster In some cases, such as when When there are multiple parallel tasks or tasks can be divided, IRS-based auxiliary transmission, Tasks can be quickly unloaded to clusters. Multiple small base stations are located within a cluster. Furthermore, several small base stations within each cluster can be selected as power sources for mobile users or radio frequency charging small base stations within the cluster.

[0067] This embodiment takes mobile users as an example. Taking the task as a divisible unit as an example, in time slots , Its tasks can be arbitrarily divided into Part of, among which For clusters The number of small base stations. Utilizing wireless communication technologies, such as NOMA, with IRS assistance, Can Each subtask is synchronously unloaded to the cluster. Inside The small base stations process the data in parallel, while the remaining data is processed locally.

[0068] S2: Calculates the time and energy required to process the task, specifically including:

[0069] S21; Calculate the time and energy of mobile users, assuming... Information transmission time is Then we can get Total time required to process the task and total energy They are respectively:

[0070]

[0071]

[0072] in They are respectively The time and energy required for local task processing for to base station Energy consumption of unloading tasks and base stations The time required to process its tasks.

[0073] S22: Calculate the energy required for the base station to process the task: Base station for The energy consumed in processing the task is:

[0074] ,

[0075] in for The rate at which tasks are unloaded, due to With base station The channel gain between the two is related to the phase and amplitude of the IRS. It is closely related to the phase and amplitude of the IRS, and It is a non-convex, non-concave function of the phase and amplitude of the IRS.

[0076] S3: With the assistance of the IRS, minimize the total energy consumption of mobile users, calculate the solution value, and perform data offloading, specifically including:

[0077] S31: In a typical scenario, with the assistance of IRS, to minimize the total energy consumption of mobile users, a task offloading model for problem P1 is constructed as follows:

[0078]

[0079] in and Phase and amplitude constraints are given for IRS in both energy acquisition and information transmission, where , The first The phase of the IRS in assisted energy harvesting and information transmission (the first) (one reflecting element), and The corresponding amplitude; and users respectively and base stations Energy constraints; For users Task execution time constraints and task quantity constraints; For users Power constraints; For base stations CPU constraints for task execution.

[0080] S32: In mobile users In scenarios where a task can only be divided into a fixed number of threads or methods, each thread or method can be considered a subtask. In this case, let... express Should subtasks be included? Unload to base station ,in express Subtasks Unload to base station ,otherwise Add the following constraints to the task unloading model for problem P1:

[0081]

[0082] in Each subtask The amount of data is And each subtask It can only be processed locally or offloaded to a small base station for processing, among which To optimize variables.

[0083] S33: In more complex scenarios such as Augmented Reality applications, mobile users The task can not only be divided into a fixed number of threads or methods, but some subtasks can only be processed locally. Furthermore, unlike problem P1, in this scenario, some subtasks have a sequential execution relationship and cannot be processed in parallel. The specific steps are as follows:

[0084] S33-1: Reference Figure 3 Using call graph To model the relationships between subtasks, the graph shows a set of vertices consisting of all subtasks, representing the dependencies between them. Subtasks 1 and 10 must be processed locally on the mobile user's device. Subtasks 2 and 3 can be processed in parallel, but subtasks 2 and 6 must be processed sequentially.

[0085] S33-2: For Assuming subtask Its predecessor task Can be used in small base stations and Local processing is required at this time, and at the base station and To pass the computation results of certain subtasks between tasks. Let Subtasks The time required for local processing They are respectively Subtasks Unload to base station Required transmission time and Processing subtasks Time required for Subtasks Preceding tasks The calculation results are offloaded to the base station. The required time, Total time required to process the task for

[13] :

[0086] .

[0087] Similarly, let For subtasks Process the required energy locally. They are respectively Subtasks Unload to base station Required transmission energy and base station Processing subtasks Energy required for Subtasks Preceding tasks The calculation results are offloaded to the base station. The required energy, Total energy required to process the task for:

[0088] .

[0089] S4: When the MEC server at the base station handles the computational tasks of mobile user offloading, it needs to first cache the corresponding application services and related databases. Therefore, when considering the task offloading problem, it is necessary to consider the service caching problem of the base station in conjunction with it. At this time, the total energy consumption of mobile users is minimized, the solution value is calculated, and data offloading is performed, specifically including:

[0090] S41: While downloading service programs typically takes a considerable amount of time, in scenarios where IRS assists in energy harvesting and information transmission, it can significantly increase wireless transmission rates and greatly reduce service program download latency. Since the MEC server's cache space at the base station is limited, IRS technology will impact the MEC server's service cache. In this case, assume the set of service programs is... Due to base stations Since the MEC server cache space is limited, the following constraints need to be added to the task offloading model when calculating the minimum energy consumption:

[0091] ,

[0092] in , At that time, base station The MEC server caches the service program at the specified location; otherwise, it does not cache it.

[0093] S42: Reference Figure 4 In scenarios where users need to upload certain custom programs to the MEC server, or where small base stations need to transmit certain application service programs wirelessly, the steps to minimize the total energy consumption of mobile users are as follows:

[0094] S42-1: The MEC service selectively caches certain service programs based on task computation requirements, thereby reducing the latency of long-term task processing for users.

[0095] S42-2: To minimize the total energy consumption of mobile users, the task constraints in problem P1 are applied. By modifying the model and adding upload constraints for service program data, a task offloading model similar to problem P2 can be constructed:

[0096]

[0097] S5: In UD-MEC systems, due to the limited spectrum resources that need to be shared among a massive number of small base stations and mobile users, severe co-channel interference exists within the system. Therefore, by selectively using millimeter-wave or terahertz communication with abundant bandwidth resources within a portion of the coverage area and during certain time periods of small base stations, co-channel interference can be reduced. Furthermore, this solution, based on IRS technology, effectively addresses the problems of millimeter-wave and terahertz bands being highly susceptible to obstruction and severe signal attenuation due to their very high frequencies and short wavelengths, thus limiting communication range. This effectively improves the user's task offloading rate to meet the microsecond-level communication latency requirements of users in the 6G era, specifically including:

[0098] S51: In order to minimize the deployment density of small base stations and further reduce deployment costs and co-channel interference, this embodiment deploys spatially mobile base stations and uses UAVs as MEC servers to dynamically execute tasks for mobile users. UAVs can provide line-of-sight transmission paths and have the characteristics of being easy to deploy, low cost, and easy to move.

[0099] S52: Reference Figure 5 The terahertz (or millimeter wave) communication environment shown includes... One UAV is used as a MEC server. With the assistance of an IRS, for the UD-MEC system Provide services to individual users. To combat frequency-selective fading in the terahertz band, the used terahertz band is divided into... Each sub-band. The time considered is divided into... Each time slot, in each time slot , By selecting different terahertz sub-bands, ground user GUs, with the assistance of IRS, offload tasks to UAVs via the terahertz bands for processing. Each UAV completes the tasks for all mobile users by flying in the air along a certain trajectory.

[0100] S53: Since UAVs have very limited energy, in order to minimize UAV energy consumption, a model for problem P3 can be constructed with the assistance of an IRS:

[0101]

[0102] in, For each time slot Time constraints and Position constraints for UAVs Unload constraints for the task. For power and energy constraints for mobile users, Assign constraints to sub-bands. Assign constraints to the CPU of the UAV.

[0103] The above content is merely an embodiment of the present invention. Commonly known structures and characteristics of the solutions are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can improve and implement this solution based on the guidance provided in this application and their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A task offloading method for an IRS-assisted radio frequency charging UD-MEC system, characterized in that: S1: Divide mobile users and base stations into several clusters; S2: Calculates the time and energy required to process the task; S2 includes: S21; Calculate the time and energy consumption of mobile users, assuming mobile users... Information transmission time is ,get Total time required to process the task and total energy They are respectively: in They are respectively The time and energy required for local task processing for to base station Energy consumption of unloading tasks and base stations The time required to process the task; S22: Calculate the energy required for the base station to process the task. For mobile users The energy consumed in processing the task is: ， in for The rate at which tasks are unloaded, due to With base station The channel gain between the two is related to the phase and amplitude of the IRS. It is closely related to the phase and amplitude of the IRS, and It is a non-convex, non-concave function of the phase and amplitude of the IRS; S3: Based on IRS assistance, minimize the total energy consumption of mobile users in several task scenarios, including regular scenarios, scenarios that can only be divided into several fixed threads, and complex scenarios that can not only be divided into several fixed threads, but also where some subtasks can only be processed locally; S3 includes: S31: In a typical scenario, to minimize the total energy consumption of mobile users, construct a task offloading model for the following problem P1: in and Phase and amplitude constraints are given for IRS in both energy acquisition and information transmission, where , The first The phase of an IRS during auxiliary energy acquisition and information transmission. and The corresponding amplitude; and For mobile users and base stations Energy constraints; For users Task execution time constraints and task quantity constraints; For users Power constraints; For base stations CPU constraints for task execution; S32: In mobile users In scenarios where a task can only be divided into a fixed number of threads, each thread can be considered a subtask. express Should subtasks be included? Unload to base station ,in express Subtasks Unload to base station ,otherwise Add the following constraints to the task unloading model for problem P1: in Each subtask The amount of data is And each subtask It can only be processed locally or offloaded to a small base station for processing, among which To optimize variables; S33: In complex scenarios, mobile users The task can not only be divided into a fixed number of threads, but some subtasks can only be processed locally. During specific processing, a call graph is used. To model the relationships between subtasks, the graph includes a set of vertices consisting of all subtasks, which is used to represent the dependencies between subtasks. Dependencies include local processing relationships, parallel processing relationships, and sequential processing relationships. S4: Jointly consider the service caching problem of base stations to minimize the total energy consumption of mobile users; S5: Selectively use millimeter wave and terahertz frequency bands to deploy UAVs as mobile base stations and minimize UAV energy consumption.

2. The task offloading method for an IRS-assisted radio frequency charging UD-MEC system according to claim 1, characterized in that: In S1, mobile users in each cluster can communicate with all small base stations within that cluster, and each cluster selects one small base station as the cluster head.

3. The task offloading method for an IRS-assisted radio frequency charging UD-MEC system according to claim 1, characterized in that: Several small base stations are selected in each cluster as the energy source for mobile users or radio frequency charging small base stations within the cluster.

4. The task offloading method for an IRS-assisted radio frequency charging UD-MEC system according to claim 1, characterized in that: S4 includes: S41: Let the set of service programs be... Due to base stations Since the MEC server cache space is limited, the following constraint is added to the task offloading model when calculating the minimum energy consumption: ， in , At that time, base station The MEC server caches the service program at the specified location, and vice versa; S42: Minimize the total energy consumption of mobile users in scenarios where users need to upload service programs to the MEC server, or where small base stations need to transmit service programs wirelessly.

5. A task offloading method for an IRS-assisted radio frequency charging UD-MEC system according to claim 4, characterized in that: S42 includes: S42-1: Selective caching service program based on task computation requirements; S42-2: To minimize the total energy consumption of mobile users, construct a task offloading model for the following problem P2: Among them, the task quantity constraints in problem P1 Modifications were made to add constraints on the uploading of service program data.

6. A task offloading method for an IRS-assisted radio frequency charging UD-MEC system according to claim 5, characterized in that: S5 includes: S51: Deploys spatially mobile base stations and uses UAVs as MEC servers to dynamically execute tasks for mobile users; S52: Divide the terahertz frequency band used into Each sub-band, time divided into In each time slot, the user selects a different terahertz sub-band, and the ground user, with the assistance of the IRS, offloads the task to the UAV for processing via the terahertz band; S53: To minimize the energy consumption of the UAV, construct a model for problem P3 as follows: in, For each time slot Time constraints and Position constraints for UAVs Unload constraints for the task. For power and energy constraints for mobile users, Assign constraints to sub-bands. Assign constraints to the CPU of the UAV.

Images