Research On Edge Computing Offloading Scheme For Smart Home

This thesis mainly discusses the application of edge computing in the smart home scene,and puts forward two kinds of joint optimization of energy consumption and computing resource cost and low delay optimization.The effectiveness of the two schemes is analyzed by comparing the simulation results.These two schemes are of great significance to the application of edge computing in smart home.First of all,it avoids the time consumption of data transmission and improves the response speed of smart home system.Moreover,computing power is concentrated on the device side to make user data more secure and reliable and avoid potential network security risks.Finally,smart homes can respond to users’ needs in real time and provide more customized intelligent services through the computing power of devices to better meet users’ needs.This thesis studies as follows:1.Study the application of edge computing in the smart home scenario of smart home.In order to reduce the cost of resource use and system energy consumption,a new task unloading scheme for smart home is proposed.This scheme aims to minimize the total system cost and jointly optimizes the energy consumption and the computing resource leasing cost under the delay constraint.This scheme uses the joint weighting method to establish the relationship between energy consumption and lease cost,and expresses it as the total target value,aiming to minimize the value.The total target value optimization problem is a NP-hard problem.In order to solve this problem,a new DEPSO algorithm is proposed.The proposed DEPSO algorithm is compared with PSO,IPSO and LISPSO algorithms of other schemes.The simulation results show that compared with other similar schemes,the proposed scheme reduces the cost and proves its effectiveness.2.Explore the application of edge computing in intelligent scene,and propose a low-delay edge computing task unloading scheme.In this scheme,the tasks generated by multiple devices are divided into local computing subtasks and unloading computing subtasks.Unloading subtasks are queued according to M/M/ m queue model.An improved hill-climbing algorithm is used to obtain the total time delay of task processing to determine the appropriate unloading ratio and minimize the time delay.The delays obtained by this scheme are compared with those obtained by local calculation and full offload calculation.The simulation results show that the delay obtained by this scheme is smaller than that obtained by local calculation and full unloading calculation,which verifies the effectiveness of this scheme.