Research On Robust Edge Computing Task Optimization Method For Channel Information Error

In recent years,with the rapid growth of the number of mobile smart devices,a large number of task processing requirements have emerged,which has also brought great challenges to the existing cloud computing system.Mobile Edge Computing(MEC)is an emerging network architecture.Decentralizing the functions of the cloud platform to the edge of the network and providing terminal devices with near-localized computing services near the user can effectively reduce task delivery delays and device energy consumption.Task offloading is one of the core issues of MEC,which is related to the overall planning of resources of the entire system.This article focuses on the task offloading decision-making in the MEC system.Taking into account the channel information error,some active studies have been carried out.Specific work as follows:1.The theoretical models commonly used in current MEC task offloading related research are summarized and organized,including the network structure,communication methods,and computational task modeling in task offloading scenarios,and the construction of energy consumption and delay during task execution The model process is derived in detail,so as to use the energy consumption and delay as optimization criteria to construct the optimization problem.2.For the task offloading scenario in an independent MEC unit,a mathematical model for data transmission and task execution is designed.The weighted sum of the overall system energy consumption and execution delay is taken as the total cost of the system,and channel information is introduced in the data transmission process.Error,a MEC unloading decision optimization problem is constructed.To solve this problem,an MEC offloading decision algorithm based on ergodic stochastic optimization is proposed.It does not need to know the probability distribution of the channel parameters after introducing the error in advance,but by observing the parameter status during each iteration.Determine the resource allocation strategy,and it can almost converge to the optimal value point.3.Extend the task offloading of independent units to multi-unit application scenarios.Considering the interference between adjacent units that transmit data through the same channel,a mathematical model of MEC task offloading is designed.An optimization problem of offloading decision is constructed.A multi-unit MEC unloading algorithm based on ergodic stochastic optimization is proposed,and the sub-problems in the iterative process are split and solved according to the unloading decision,and the result is approximated to the optimal value point by iteration.Computer simulation verification of the above work,with the introduction of channel information error,compared with the conventional algorithm,the proposed algorithm can give a smaller total system cost,and with the increase in the number of equipment and channel information error,the proposed algorithm can still maintain good performance.