Research On Load Balancing Algorithm In Heterogeneous Integrated Wireless Networks Environment

With the development of wireless communication technology and the emergence of new radio access technology, the integration of heterogeneous wireless networks will be one of the main features of the next generation wireless communication system. Load balancing as a key techniques of RRM can improve the utilization of system wireless resource and Qos, avoid the deterioration of system performance. Therefore it is of great theoretical value and practical significance for the research of load balancing algorithm.The thesis introduces the basic theory of integrated heterogeneous networks, elaborates three key technologies of load balancing, lucubrates and analyzes the advantages and disadvantages of the existing algorithms for load balancing in integrated heterogeneous network, then propose two effective and applicable load balancing algorithm based on these works.A Load balancing algorithm based on load forecasting is proposed from the communication services access control point of view in heterogeneous integrated wireless networks environment. The algorithm utilizes dynamic load forecasting algorithm based on mutual information and BP neural network to obtain real time load rate of each network in system, and makes load balancing as a final target for satisfying the condition of guaranteeing the quality of service for communication services. Then an optimization model is built, and the optimal solution for services access control is obtained by genetic algorithm. Simulation results show that the algorithm proposed in this thesis has a better performance in both reducing the system average traffic blocking rate and improving network load balancing degrees compared with the load balancing algorithm based on multi-access selection.A Load balancing algorithm based on general utility function is proposed from the load transfer point of view in heterogeneous integrated wireless networks environment. The algorithm designs two utility functions for real time network load rate with dynamic blocking rate of RT services and real time network load rate with average transmission time of NRT services separately, then fuses the two utility functions by triangular norm operator to obtain the comprehensive performance evaluation of candidate access networks, and transfer an appropriate amount of services load from heavy load network to light load network. Simulation results show that the algorithm proposed in this thesis has a better performance in ensuring quality of service, system resource utilization rate, and the load balancing degree has been improved compared with both the load balancing algorithm based on multi-access selection and the dynamic load balancing algorithm discussed in this thesis.