Research On Intelligent Home Intrusion Detection Based On Alo Optimized SVM

In today's era,Internet technology has developed rapidly.In order to adapt to the development speed and direction of the times,computer networks are also constantly popularizing and living,and are widely used to exchange confidential data information between servers and mobile devices and desktops.Major IoT organizations and smart home manufacturers have proposed their own solutions.Different standards and protocols bring about the diversification of smart home products,but also because of the inconsistency of standards and protocols,which brings many problems.The standards and protocols proposed by major IoT organizations and smart home manufacturers generally do not pay enough attention to security issues.Therefore,it is of practical and practical value to study smart home systems with compatibility and high security.This paper designed a smart home system based on Alljoyn's thin client for the above problems.This system consists of three parts: standard client,device system bridge and thin client.The experimental results show that the system can discover,access,configure,and generate a unified control interface and control operations for thin client devices through standard client applications.Then,an intrusion detection method is introduced for the security problem of the potential network attack of the smart home system.The first is the choice of classifier.In recent years,machine learning has become an important part of the field of network security and intrusion detection,and many algorithms have been generated to solve various problems.But which of these algorithms will enhance the intrusion detection system that will solve the smart home environment has become the problem to be solved in this paper.This paper proves that the Support Vector Machine(SVM)is suitable for the smart home environment.According to the selected classifier,an intrusion detection method based on ant lion optimization algorithm to optimize SVM is proposed.The ant lion optimization algorithm is used to optimize the SVM parameters,and the optimal penalty factor and kernel function are obtained to establish the optimal classification model,so as to improve the classification accuracy of network intrusion detection.The intrusion detection method is part of data collection and preprocessing,optimization of parameters,training and learning.In this paper,the ant optimization algorithm is used to test the parameters of SVM.The selection test of the classifier and the intrusion detection method based on the ant lion optimization algorithm to optimize the SVM are all performed on the UNSW-NB15 network data set.First,the classifier selection experiment was to select five classifiers for experiments and then use five evaluation criteria for evaluation.The results show that SVM is more suitable for smart home systems.Finally,based on the ant lion optimization algorithm,the SVM intrusion detection method and the particle swarm optimization algorithm are optimized,and the optimization algorithm is compared experimentally.In these experiments,the ant lion optimization algorithm optimizes the SVM classification network intrusion detection method to have higher efficiency and detection rate based on the reduction of false positive rate.