Internet of vehicles task unloading method based on optimal stopping theory

Abstract

The invention relates to a calculation task optimization unloading method and system in the Internet of Vehicles based on the optimal stop theory, and belongs to the technical field of the Internet of Vehicles, and the method comprises the steps: constructing an Internet of Vehicles model based on the optimal stop theory, and building a target function for minimizing the total unloading cost expectation according to the vehicle speed and the RSU load; the cost data comprises the load and observation cost of the RSU; solving the objective function by adopting an optimal stopping theory, and calculating to obtain a reference value; according to the observed current RSU load, the current load is compared with the reference value obtained through calculation, and the optimal stop moment is found for unloading. According to the method, the calculation tasks are optimized and unloaded, so that load balancing of task calculation in the Internet of Vehicles is realized, and the utilization rate of calculation resources is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle networking, and in particular to a method for unloading tasks in a vehicle networking based on optimal stopping theory. Background Art

[0002] The computing task offloading strategy in the Internet of Vehicles is a hot research direction. Because the computing power of the vehicle itself is limited, it often needs to communicate with the roadside unit (RSU) to offload computing-intensive tasks. Because different edge servers have different loads, the vehicle wants to offload data to an RSU with a lower load, so the vehicle can use its mobility advantage to observe the edge server. In order to help the vehicle choose the RSU with the least load as much as possible for task offloading, it is necessary to propose a task offloading method based on the optimal stopping theory to decide when the vehicle should unload. Summary of the invention

[0003] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and to provide a vehicle networking task unloading system and method based on the optimal stopping theory. By optimizing the unloading of computing tasks, the load balancing of task computing in the vehicle networking is achieved, thereby improving the utilization rate of computing resources.

[0004] According to one aspect of the present invention, the present invention provides a method for optimizing unloading of computing tasks in a connected vehicle network based on optimal stopping theory, the method comprising the following steps:

[0005] S1: Construct a vehicle networking model based on the optimal stopping theory, and establish an objective function that minimizes the total unloading cost expectation according to the vehicle speed and the load of the RSU; the total unloading cost includes the load of the RSU and the observation cost;

[0006] S2: solving the objective function using the optimal stopping theory to calculate a reference value;

[0007] S3: According to the observed current RSU load, the current load is compared with the calculated reference value to find the optimal stop time for unloading.

[0008] Preferably, the objective function of establishing the expected minimization of the offloading cost includes:

[0009] The objective function of vehicle unloading is established and the optimization problem is as follows:

[0010]

[0011] in, represents the total unloading cost of vehicle i, x nrepresents the load of server n, that is, the total delay spent on offloading tasks to the selected server for processing; ΔL represents the distance between adjacent RSUs, n is the number of the current RSU, and v i is the driving speed of vehicle i, and c represents the observation cost required to observe an RSU.

[0012] Preferably, the optimal unloading cost of vehicle i at its nth observation is

[0013]

[0014] in, The offloading cost of stopping observation and selecting the current server at this time is, is the expected cost of continuing the observation.

[0015] Preferably, if vehicle i has not been unloaded until the end of observation, it can only choose to unload to the last RSU. The optimal unloading cost of the vehicle at the Nth observation is

[0016]

[0017] in, is the offloading cost incurred by offloading to the Nth server.

[0018] Beneficial effect: The present invention constructs a vehicle network model based on the optimal stopping theory, establishes an objective function that minimizes the expected total unloading cost according to the vehicle speed and the load of the RSU, so as to find the optimal RSU for task unloading. The optimal stopping theory is used to solve this problem, and the observed current RSU load is compared with a calculated scalar value to determine the optimal unloading time, thereby achieving load balancing of RSUs in the vehicle network and improving the efficiency of task unloading.

[0019] The features and advantages of the present invention will become clear through reference to the following drawings and detailed description of specific embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of the method for optimizing offloading of computing tasks in the Internet of Vehicles based on the optimal stopping theory;

[0021] Figure 2 It is a cost comparison chart of the proposed algorithm, random algorithm and optimal algorithm. DETAILED DESCRIPTION

[0022] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] Figure 1 This is a flow chart of the method for optimizing the offloading of computing tasks in the Internet of Vehicles based on the optimal stopping theory. Figure 1 As shown, this embodiment provides a method for optimizing unloading of computing tasks in a connected vehicle network based on the optimal stopping theory, and the method includes the following steps:

[0024] In order to find the optimal RSU for task offloading, the optimal stopping theory is used to solve this problem. The observed current RSU load is compared with one to determine the optimal unloading time, so as to achieve load balancing of RSUs in the Internet of Vehicles and improve the efficiency of task offloading.

[0025] S1: Construct a vehicle networking model based on the optimal stopping theory, and establish an objective function that minimizes the total unloading cost expectation according to the vehicle speed and the load of the RSU; the total unloading cost includes the load of the RSU and the observation cost;

[0026] S2: solving the objective function using the optimal stopping theory to calculate a reference value;

[0027] S3: According to the observed current RSU load, the current load is compared with the calculated reference value to find the optimal stop time for unloading.

[0028] The method of this embodiment optimizes and unloads computing tasks, thereby achieving load balancing of task computing in the Internet of Vehicles and improving the utilization rate of computing resources.

[0029] In step S1, the objective function of minimizing the expected offloading cost is established including:

[0030] The objective function of vehicle unloading is established and the optimization problem is as follows:

[0031]

[0032] in, represents the total unloading cost of vehicle i, x n represents the load of server n, that is, the total delay spent on offloading tasks to the selected server for processing; ΔL represents the distance between adjacent RSUs, n is the number of the current RSU, and v i is the driving speed of vehicle i, and c represents the observation cost required to observe an RSU.

[0033] Specifically, a network consisting of I vehicle terminals and N RSUs is constructed. N RSUs equipped with edge servers are deployed on the roadside, and each vehicle travels at a speed v i Driving on the road. The vehicle generates a computing task that needs to be unloaded and starts to observe the RSUs in sequence. The whole process can be divided into N observations. The optimal unloading cost of vehicle i at its nth observation is

[0034]

[0035] in, The offloading cost of stopping observation and selecting the current server at this time is, is the expected cost of continuing the observation.

[0036] If vehicle i has not been unloaded until the end of observation, it can only choose to unload to the last RSU. The optimal unloading cost of the vehicle in the Nth observation is

[0037]

[0038] in, is the offloading cost incurred by offloading to the Nth server.

[0039] In step S2, the optimal stopping theory is used to solve the objective function to obtain the parameter ξ, and the specific process is as follows:

[0040] In step S2, to select the optimal value as much as possible, the vehicle needs to unload when the current unloading cost is lower than the expected future unloading cost:

[0041]

[0042] in, represents the total unloading cost of vehicle i, x n represents the load of server n, that is, the total delay spent on offloading tasks to the selected server for processing; ΔL represents the distance between adjacent RSUs, n is the number of the current RSU, and v i is the driving speed of vehicle i, and c represents the observation cost required to observe an RSU.

[0043] Rearranging the above formula, we can know that x n The following conditions must be met:

[0044]

[0045] For simplicity and subsequent convenience, a

[0046]

[0047] At this point, you only need to change x n and For comparison, Substitute to calculate

[0048]

[0049] Here the random sequence x n Convert to unknown number x:

[0050]

[0051] Where f(x) represents the probability density function of the variable x, and F(x) represents the cumulative distribution function of the variable x.

[0052] In step S3, the reference value is calculated through the above process is compared with the RSU load currently observed by the vehicle. If At this time, the optimal strategy is to stop observing and offload the task to the current server; if The best thing to do at this time is to continue observing.

[0053] Figure 2 is a cost comparison diagram of the proposed algorithm, the random algorithm, and the optimal algorithm. As can be seen from the figure, compared with the random unloading method, the method proposed in this embodiment can achieve a lower cost. The algorithm proposed in this article can achieve a cost close to that of the optimal algorithm.

[0054] This embodiment constructs a vehicle network model based on the optimal stopping theory, establishes an objective function that minimizes the expected total unloading cost according to the vehicle speed and the load of the RSU, solves the optimal computing task unloading method, and uses the optimal stopping theory to solve this problem. The observed current RSU load is compared with a calculated scalar value to determine the optimal unloading time, thereby achieving load balancing of RSUs in the vehicle network and improving the utilization of computing resources.

[0055] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A method for optimizing unloading of computing tasks in a connected vehicle network based on optimal stopping theory, characterized in that: The method comprises the following steps: S1: Construct a connected vehicle model based on the optimal stopping theory, and establish an objective function that minimizes the total unloading cost expectation according to the vehicle speed and the load of the RSU; S2: solving the objective function using the optimal stopping theory to calculate a reference value; S3: According to the observed current RSU load, the current load is compared with the calculated reference value to find the optimal stop time for unloading.

2. The method for optimizing unloading of computing tasks in the Internet of Vehicles based on the optimal stopping theory according to claim 1 is characterized in that: The total offloading cost includes the load of the RSU and the observation cost.

3. The method for optimizing unloading of computing tasks in the Internet of Vehicles based on the optimal stopping theory according to claim 1 is characterized in that: The objective function to minimize the expected computational cost is established by: The objective function of vehicle unloading is established, and the optimization problem formula is as follows: in, represents the total unloading cost of vehicle i, x n represents the load of server n, that is, the total delay spent on offloading tasks to the selected server for processing; ΔL represents the distance between adjacent RSUs, n is the number of the current RSU, and v i is the driving speed of vehicle i, and c represents the observation cost required to observe an RSU.

4. The method for optimizing unloading of computing tasks in a connected vehicle network based on optimal stopping theory according to claim 3 is characterized in that: The optimal unloading cost of vehicle i at its nth observation is: in, The offloading cost of stopping observation and selecting the current server at this time is, is the expected cost of continuing the observation.

5. The method for optimizing unloading of computing tasks in the Internet of Vehicles based on the optimal stopping theory according to claim 3 is characterized in that: If vehicle i has not been unloaded until the end of observation, it chooses to unload to the last RSU. The optimal unloading cost of the vehicle in the Nth observation is: in, is the offloading cost incurred by offloading to the Nth server.

Images