A method for optimizing energy consumption of task offloading for non-uniform speed vehicle users

Abstract

The present invention claims protection for a method for optimizing the energy consumption of task offloading for non-uniform speed vehicle users, and belongs to the field of communication technology. In view of the uncertainty of the residence time of non-uniform speed vehicle users in a cell, which makes it difficult for the planned task offloading scheduling to meet the task tolerance delay requirements and deteriorates the energy efficiency of the user end, a periodic task offloading scheduling method based on task tolerance delay and V2I link duration prediction is proposed. This method predicts the duration of the V2I link based on the real-time speed and location information of the vehicle, and periodically plans the unloading task scheduling of the vehicle user based on the task tolerance delay. By reasonably selecting and allocating the edge server and the user's local computing resources, the task tolerance delay requirements are met to the maximum extent and the unloading energy consumption of the vehicle user is minimized.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a method for optimizing energy consumption of task offloading for non-uniform speed vehicle users. Background Art

[0002] With the continuous development of IoT technology, a large number of new applications and services for user terminals have emerged, such as augmented reality, autonomous driving, and voice recognition. These applications and services usually have high latency requirements and require a lot of computing resources and energy consumption. Since cloud servers are often deployed far away from user terminals, they will cause high propagation delay overhead. Deploying edge servers in roadside units (RSUs) of edge networks can effectively reduce user task offloading latency and save energy consumption of terminal devices.

[0003] Existing solutions for vehicle mobile edge computing mainly focus on the task offloading scheduling problem under the condition of uniform vehicle speed. However, in reality, the uncertainty of the retention time of non-uniform vehicle users in the community can easily lead to unreasonable allocation of local computing resources between the server and the user. The planned task offloading scheduling is difficult to meet the task tolerance delay requirements and deteriorates the energy efficiency of the user end. To address this problem, a periodic task offloading scheduling method based on task tolerance delay and V2I link duration prediction is proposed. According to the real-time vehicle speed and location information, the task offloading scheduling is periodically planned, as well as the selection and allocation of resources, to maximize the task tolerance delay requirements and minimize the task offloading energy consumption of vehicle users. Summary of the invention

[0004] The present invention aims to solve the above problems of the prior art. A method for optimizing the energy consumption of task offloading for non-uniform speed vehicle users is proposed. The technical solution of the present invention is as follows:

[0005] A method for optimizing energy consumption of task offloading for non-uniform speed vehicle users comprises the following steps:

[0006] 101. Initiate task unloading request D based on vehicle j entering community i j , get the size s of the corresponding task in the task offloading request j , task complexity c j , task tolerance delay Initialize the amount of completed tasks Cycle number k = 0, binary variable

[0007] 102. Let k = k + 1, According to the current speed v of vehicle j j and location information Calculate the kth unloading period τ j，kand V2I link duration According to the amount of completed tasks and delay tolerance Update the computational resource requirements f of vehicle j j，k ;

[0008] 103. If According to f j，k and edge server computing resource status, planning τ j，k The amount of tasks offloaded to edge servers And allocate edge server computing resources to vehicle j Jump to step 104, otherwise, let Jump to step 104;

[0009] 104. According to and the local computing resource status of vehicle j, planning τ j，k The amount of tasks executed locally and allocate local computing resources Start Execution and offloading of tasks;

[0010] 105. If within period k and If the task uninstallation result is not returned, jump to step 106. Otherwise, Jump to step 107;

[0011] 106. If cycle k has not ended, jump to step 105; otherwise, jump to step 107;

[0012] 107. Release server computing resources and local computing resources If vehicle j switches from cell i to cell i+1, let i=i+1 and jump to step 108; otherwise, jump to step 108;

[0013] 108. Order if Jump to step 102, otherwise, jump to step 109;

[0014] 109. The algorithm ends.

[0015] Furthermore, the uninstallation period τ in step 102 j，k The calculation method is shown in formula (1):

[0016]

[0017] In formula (1), variables α∈(0, 10], β∈(0, 1] are used to adjust the current vehicle speed vj and unloading period τ j，k , τ0 is the minimum unloading period;

[0018] The duration of the V2I link in step 102 The calculation method is shown in formula (2):

[0019]

[0020] In formula (2), the position information Represents the distance from the current position of vehicle j to the exit of community i.

[0021] Furthermore, in step 102, the resource requirement f is calculated j,k The calculation method is shown in formula (3):

[0022]

[0023] In formula (3), Indicates the time spent on executing the task, r i,j represents the V2I transmission rate, f i Indicates the remaining available resources of the server in cell i.

[0024] Furthermore, in step 103, τ is planned j,k The amount of tasks offloaded to edge servers The calculation method is shown in formula (4):

[0025]

[0026] Furthermore, in step 103, edge server computing resources are allocated The calculation method of is shown in formula (5):

[0027]

[0028] Furthermore, in step 104, τ is planned j,k The amount of tasks executed locally The calculation method of is shown in formula (6):

[0029]

[0030] In formula (6), f j is the remaining available computing resources locally for vehicle j.

[0031] Furthermore, in step 104, local computing resources are allocated The calculation method of is shown in formula (7):

[0032]

[0033] The advantages and beneficial effects of the present invention are as follows:

[0034] The present invention discloses a method for optimizing the energy consumption of task offloading for non-uniform speed vehicle users. Existing solutions for vehicle mobile edge computing mainly focus on the task offloading scheduling problem under the condition of uniform vehicle speed. However, in reality, the uncertainty of the residence time of non-uniform speed vehicle users in the community can easily lead to unreasonable allocation of local computing resources between the server and the user. The planned task offloading scheduling is difficult to meet the task tolerance delay requirements and deteriorates the energy efficiency of the user end. The present invention predicts the duration of the V2I link based on the real-time vehicle speed and location information, and dynamically adjusts the offloading cycle. According to the V2I link duration and the task tolerance delay, the edge server and the user's local computing resource allocation are optimized to maximize the task tolerance delay requirements and minimize the vehicle user's offloading energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 The present invention provides a preferred embodiment of a method for optimizing the energy consumption of task unloading for non-uniform speed vehicle users. DETAILED DESCRIPTION

[0036] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.

[0037] The technical solution of the present invention to solve the above technical problems is:

[0038] The concepts and models involved in the content of the present invention are as follows.

[0039] 1. Network Model

[0040] The study of the vehicle networking scenario with real-time changes in vehicle speed consists of vehicles, roadside base stations RSU and edge servers. An edge server with limited computing resources is deployed on each RSU, and its service scope is all users in the current cell. The vehicle-mounted equipment can perform some tasks locally, or offload some tasks to the edge server of the current cell for parallel execution.

[0041] 2. Other symbols involved in the present invention are explained as follows.

[0042] i: cell i

[0043] j: vehicle j

[0044] k: period k

[0045] s j : Vehicle j mission size

[0046] cj :The computational complexity of vehicle j task

[0047] Tolerance delay of vehicle j task

[0048] The amount of tasks completed by vehicle j

[0049] Determine whether it is a valid unloading cycle flag

[0050] v j : Real-time speed of vehicle j

[0051] The distance from the current position of vehicle j to the exit of community i.

[0052] τ j,k : The kth unloading cycle of vehicle j

[0053] V2I link duration

[0054] r i,j : V2I link transmission rate

[0055] f j,k : Computational resource requirements for vehicle j task

[0056] f i : Remaining available computing resources of the server in cell i

[0057] Server computing resources allocated by cell i to vehicle j

[0058] f j : The remaining available computing resources in the local area of ​​vehicle j

[0059] The computational resources allocated locally to vehicle j

[0060] The time spent on executing vehicle j task

[0061] Vehicle j planning τ j,k The amount of tasks offloaded to edge servers

[0062] Vehicle j planning τ j,k The amount of tasks executed locally

[0063] The technical solution of the present invention is described as follows.

[0064] 1. Unloading cycle τ j,k

[0065] The calculation method is shown in formula (1):

[0066]

[0067] In formula (1), variables α∈(0,10], β∈(0,1] are used to adjust the current vehicle speed v j and unloading period τ j,k , τ0 is the minimum unloading period.

[0068] 2. V2I link duration

[0069] The calculation method is shown in formula (2):

[0070]

[0071] 3. Computational resource requirements j,k

[0072] The calculation method is shown in formula (3):

[0073]

[0074] 4. Planning τ j,k The amount of tasks offloaded to edge servers

[0075] The calculation method is shown in formula (4):

[0076]

[0077] 5. Allocate edge server computing resources

[0078] The calculation method is shown in formula (5):

[0079]

[0080] 6. Planning τ j,k The amount of tasks executed locally

[0081] The calculation method is shown in formula (6):

[0082]

[0083] 7. Allocate local computing resources

[0084] The calculation method is shown in formula (7):

[0085]

[0086] A method for optimizing energy consumption of task offloading for non-uniform speed vehicle users, wherein the specific implementation method includes the following steps.

[0087] Step 1: Initiate a task offloading request D based on vehicle j entering community i j , get the size s of the task j , task complexity c j , task tolerance delay Initialize the amount of completed tasks Cycle number k = 0, binary variable

[0088] Step 2: Let k = k + 1, According to formula (1), the kth unloading period τ is calculated j,k , calculated according to formula (2) According to formula (3), f is calculated j,k ;

[0089] Step 3: If According to formula (4), According to formula (5), Jump to step 4, otherwise, let Jump to step 4;

[0090] Step 4: Calculate according to formula (6) According to formula (7) Start Execution and offloading of tasks;

[0091] Step 5: If within period k and If the task uninstallation result is not returned, jump to step 106. Otherwise, Jump to step 107;

[0092] Step 6: If cycle k has not ended, jump to step 5, otherwise, jump to step 7;

[0093] Step 7: Release server computing resources and local computing resources If vehicle j switches from cell i to cell i+1, let i=i+1 and jump to step 8; otherwise, jump to step 8;

[0094] Step 8: Order if Jump to step 2, otherwise, jump to step 9;

[0095] Step 9: The algorithm ends.

[0096] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0097] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A method for optimizing energy consumption of task offloading for non-uniform speed vehicle users, characterized in that: The following steps are involved:

101. Initiate task unloading request D based on vehicle j entering community i j , get the size s of the corresponding task in the task offloading request j , task complexity c j , task tolerance delay Initialize the amount of completed tasks Cycle number k = 0, binary variable 102. Let k = k + 1, According to the current speed v of vehicle j j and location information Calculate the kth unloading period τ j,k and V2I link duration According to the amount of completed tasks and delay tolerance Update the computational resource requirements f of vehicle j j,k ; 103. If According to f j,k and edge server computing resource status, planning τ j,k The amount of tasks offloaded to edge servers And allocate edge server computing resources to vehicle j Jump to step 104, otherwise, let Jump to step 104; 104. According to and the local computing resource status of vehicle j, planning τ j,k The amount of tasks executed locally and allocate local computing resources Start Execution and offloading of tasks; 105. If within period k and If the task uninstallation result is not returned, jump to step 106. Otherwise, Jump to step 107; 106. If cycle k has not ended, jump to step 105; otherwise, jump to step 107; 107. Release server computing resources and local computing resources If vehicle j switches from cell i to cell i+1, let i=i+1 and jump to step 108; otherwise, jump to step 108; 108. Order if Jump to step 102, otherwise, jump to step 109; 109. The algorithm ends; The unloading period τ in step 102 j,k The calculation method is shown in formula (1): In formula (1), variables α∈(0,10], β∈(0,1] are used to adjust the current vehicle speed v j and unloading period τ j,k , τ0 is the minimum unloading period; The duration of the V2I link in step 102 The calculation method is shown in formula (2): In formula (2), the position information represents the distance from the current position of vehicle j to the exit of community i; In step 102, the resource requirement f is calculated. j,k The calculation method is shown in formula (3): In formula (3), Indicates the time spent on executing the task, r i,j represents the V2I transmission rate, f i Indicates the remaining available resources of the server in cell i; In step 103, τ is planned j,k The amount of tasks offloaded to edge servers The calculation method is shown in formula (4): In step 103, edge server computing resources are allocated The calculation method of is shown in formula (5): In step 104, τ is planned j,k The amount of tasks executed locally The calculation method of is shown in formula (6): In formula (6), f j The remaining available computing resources of vehicle j; In step 104, local computing resources are allocated The calculation method of is shown in formula (7):

Images