Network Selection Algorithms Based On Machine Learning In Heterogeneous Wireless Networks

With the development of communication technology,a variety of heterogeneous wireless networks overlap and cover each other.They have their own strengths and can provide users with many kinds of services.With the continuous development of technology and the gradual evolution of the networks,all kinds of existing and future emerging wireless access technologies coexist.They complement and compete with each other to form the next generation of ubiquitous wireless heterogeneous networks.In this case,how to design an intelligent and efficient network selection algorithm is a key research topic.Firstly,this thesis improves the traditional multi-attribute decision-making algorithm,and proposes a network selection algorithm based on AHP and similarity degree.Then combining machine learning with network selection,a network selection algorithm based on decision tree is put forward.Finally,a network access decision algorithm based on weighted GRA and genetic algorithm is proposed,which combines traditional multi-attribute decision-making and genetic algorithm.The main contents of this thesis are as follows.(1)This thesis proposes a network selection algorithm based on AHP and similarity degree.The services are divided into three: Conversational Class,Streaming Class and Interactive Class.According to the characteristics of each service,a different judgment matrix is given and then the analytic hierarchy process(AHP)method is used to calculate the network attributes weight.Taking the dynamic changes in user demands and network environment into account,a formula based on Lance distance for computing the attributes similarity is derived to evaluate the degree of conformity between user requirements and network attributes,from which the similarity between the user requirements and network attributes is calculated and then the total similarity by weighting.The network with the largest total similarity is the best choice.Simulation results demonstrate the effectiveness of our scheme in improving the quality of service according to the user requirements under three kinds of services.(2)This thesis puts forward a network selection algorithm based on decision tree.First,we get the training data under three kinds of services from the synergetic algorithm which can be used for training set.The network attributes are used for attribute set.And then we can choose the attribute with the largest information gain as the division attribute after the discretization of continuous features by the bisection method.Keep going this step recursively,we can finally get a decision tree with high generalization ability by which we can make the network selection.Simulation indicate that the algorithm we put forward is easy and valid and demonstrate the effectiveness of our scheme in improving the quality of service according to the user requirements under three kinds of service.(3)This thesis proposes a network access decision algorithm based on weighted GRA and genetic algorithm.Firstly,the hierarchical structure of network selection problem is established by using AHP analytic hierarchy process,and the judgment matrix is established to obtain the subjective weights of network attributes.Then this thesis combines the genetic algorithm with grey relational analysis method to define the fitness function in genetic algorithm,and it applies the selection operator,crossover operator and mutation operator in genetic algorithm to the initial weight.Constantly iterating to get the most adaptable network as the target network,which can effectively improve the quality of user service.