Research On Heterogeneous Private Network Selection Algorithm Based On Multiattribute Fuzzy Decision

At present,China's police mobile communication network mainly includes narrowband Police Digital Trunking(PDT)and Broadband Trunking Communication(B-Trun C).With the rapid development of wireless communication technology and smart terminal,public security private network communication is no longer only satisfied with the basic voice communication service,but has generated more abundant business requirements.The heterogeneous network of PDT+B-Trun C is the future development direction of the private network.Network selection problem is the main problem that must be solved in heterogeneous networks.This paper proposes the heterogeneous private network selection algorithm ba sed on Multi-Attribute Decision Making(MADM)and Fuzzy Logic(FL)respectively,which can better solve the network selection problem and have some advantages compared with other algorithms.First,this paper introduces the advanced mobile police intelligent fusion network architecture and heterogeneous private network system model.According to the characteristics of private network communication,the services are divided into four types: trunking voice service,trunking video service,trunking supplementary service and short data service.And constructed a PDT and B-Trun C heterogeneous private network model.According to the frequency characteristics of the network,the appropriate channel transmission model is selected to calculate the RSS of the network.Secondly,a MADM algorithm based on comprehensive weight and relative entropy TOPSIS is proposed.The AHP and the CRITIC are used to calculate the subjective and objective weights of each network attribute,respectively,and combined into a comprehensive weight.Relative entropy is introduced to replace the distance between Euclidean distance calculation schemes.And the algorithm is compared with the traditional TOPSIS algorithm and SAW algorithm for simulation comparison and analysis.Finally,the fuzzy theory is applied to the network selection algorithm.A network selection algorithm based on parameter fuzzy and MADM is proposed.The candidate network which can not meet the mobile speed of the terminal is eliminated by the speed pre-decision module.Then some uncertain network attributes are fuzzified by using the fuzzy theory,and the terminal speed and network attrib utes are considered comprehensively when calculating the network evaluation value.The simulation results show that the algorithm has better performance than the traditional RSS-based algorithm and MADM algorithm.On this basis,a network selection algorithm based on fuzzy inference is proposed,which applies the complete process of fuzzy logic to network selection.In the fuzzification part,according to the characteristics of different business types,different membership functions are set up,then the fuzzy inference rule base is constructed,and finally the center of gravity method is used in defuzzification.The simulation results show that the algorithm is better than other algorithms.