The Research On Traffic Steering Algorithm In The Multi-bands LTE Network

At present the situation that multi-RATs and multi-bands coexistence in the same coverage will exist for a long time, and the user is likely to choose the non-optimal access network, resulting in a waste of resources. In order to make an efficient use of all available resources to carry the complete offered user traffic, this thesis studies the Traffic Steering algorithm.This thesis first briefly introduces the importance of network optimization, which leads to a new optimization approach-Traffic Steering, and introduced its current research at home and abroad. Then introduce the background of the LTE system, and to study problems of the current LTE network optimization and some solutions. Then study the background of Traffic Steering algorithms, implementations, the basic criteria, four original Traffic Steering algorithms, and introduces the simulation and studies of one Traffic Steering algorithm which is based on load and coverage.This thesis argues that the factors considered in the current TS algorithm is very limited and the method is also not flexible, therefore propose to combine the Analytic Hierarchy Process algorithm with TS algorithm. After a simple understanding of the development and research of multi-objective decision, we sort and introduce kinds of methods in accordance with the targets’number is limited or not classification. Then, we focus on Analytic Hierarchy Process (AHP), learn its strengths and weaknesses, and then study the specific steps of AHP in detail, in traffic steering using AHP to calculate the current comprehensive performance of the networks, to choose the best access network for part users who have poor performance.In this thesis, after studying the LTE system, basic TS algorithm and AHP, we build a multi-bands LTE platform with the Matlab simulation. We give the simulation modeling, statistical models and simulation processes, and then analyze from the maximum capacity of the overall system, the total average of user rate and the three total success rate of BE traffic (access success rate, drop rate, handover success rate) to verify the new TS algorithm. In the case of sacrificing a small amount of KPI, the system capacity and user’data rate have been enhanced, so the new TS algorithm is indeed feasible.