Research On Intrusion Detection For Smart Home

In recent years,the spread of 5G technology has accelerated the construction of "smart cities".Smart home industry has also been rapid development,intelligent door locks,intelligent home appliances can be seen everywhere in life,the era of intelligent family life has come.Although smart home improves quality of people’s life,it also brings many security problems to users.Intrusion detection technology,as an active security defense technology,is still in its infancy in the smart home industry.Faced with massive high-dimensional data,existing detection schemes have some problems such as low detection rate.In addition,the current smart home system based on cloud computing cannot meet the real-time requirements of security protection applications.Therefore,it is of great significance to study the smart home system with high security and low delay time.The main work of this thesis is as follows:1.Aiming at the massive high-dimensional data,this thesis selects the kernel principal component analysis algorithm(KPCA)to extract the features of the data and reduce the dimension,thus reducing the amount of data calculation.In order to achieve better effect of KPCA algorithm,this paper improves Fruit Fly Algorithm(FOA),and uses it to optimize KPCA algorithm.2.The intrusion detection model based on LFKPCA-Adacost-DWELM is proposed.Firstly,a Weighted Extreme Learning Machine(WELM)based on data dispersion(DWELM)is proposed to improve WELM.Then the Adacost algorithm is improved,and the improved Adacost algorithm and DWELM algorithm are combined as the classifier of intrusion detection model.Finally,a comparative experiment has been carried out to prove its good detection performance.3.Designing and implementing a smart home intrusion detection system.A smart home system platform based on edge computing technology framework kubeedge is built to meet the real-time requirements of smart home security applications.According to the intrusion detection model based on LFKPCA-Adacost-DWELM proposed in this thesis,the intrusion detection system is designed and implemented,and it is deployed on the edge node of the smart home system platform.The attack experiment proves that it can detect the intrusion behavior accurately and in real time.