A Method of Adaptive Joint Computing Offloading and Resource Allocation in Vehicle Networking

Abstract

The invention relates to an adaptive joint computing offloading and resource allocation method in the Internet of Vehicles, belonging to the field of edge computing in the Internet of Vehicles. The method includes considering the concurrent unloading of multiple vehicles under the combination of local, MEC server, remote cloud server, and idle vehicle multi-computing platforms. Comprehensively consider the size of the vehicle task, the maximum tolerance delay, the computing resources under the RSU, and the network bandwidth factors, and offload the computing tasks of the vehicle to multiple computing platforms; automatically adjust the offloading platform and unloading of the vehicle computing tasks according to the number of tasks under the RSU Ratio, allocate the computing resources of MEC while the vehicle obtains the optimal unloading ratio; model the computing unloading decision and resource allocation as a multi-constraint optimization problem; use the compressed particle swarm optimization algorithm, and propose a particle matrix encoding method to jointly optimize unloading decision making and resource allocation. Experimental results show that the present invention can minimize the total system cost while satisfying the maximum tolerable time delay.

Description

technical field [0001] The invention belongs to the field of Internet of Vehicles computing, and relates to an adaptive joint computing unloading and resource allocation method in the Internet of Vehicles. Background technique [0002] As the number of autonomous vehicles in C-V2X increases, various computation-intensive and latency-sensitive applications emerge, such as image-assisted navigation and augmented reality driving, which require massive computing resources for real-time processing and analysis of large amounts of transmitted data. sensory data, which poses a huge challenge for vehicles with limited computing resources. [0003] By offloading computing tasks to other computing nodes, it can effectively solve the problem of limited computing resources in vehicles. The remote cloud server has rich computing resources, but it is far away from the vehicle, which will cause huge transmission delay and energy consumption. The MEC server sinks the calculation to the ro...

AI Technical Summary

Problems solved by technology

In fact, most vehicles are unloaded concurrently. For MEC servers with limited computing resources, it is difficult to meet the maximum tolerance delay. Therefore, it is necessary to consider the unloading of multiple computing platforms and the reasonable allocation of resources. For the green Internet of Vehicles, consider Energy consumption is also very necessary

Images