Study On Key Technology Of Self-Organizing Network In LTE-A Heterogeneous Networks

With the rapid development of wireless communication technology, the parameters and structure of LTE-A network are more complex and diverse, and the wide deployment of small power wireless devices, the coexistence of multiple heterogeneous networks, and the elimination of the interference of heterogeneous networks all affect the operation of the network. At the same time, users are increasingly sensitive to the quality of service provided by the wireless network and need increasing higher performance requirements of the network. The network must meet the needs of users to ensure the quality of service and user experience, so as to ensure the user’s loyalty and reduce the loss of users. In order to reduce manual intervention, maximize network operation efficiency, reduce infrastructure costs and network operation costs, LTE 3GPP standard introduces SON technology. SON includes self-configuration, self-optimization and self-healing. The main purpose of SON is that the wireless network can achieve some independent functions, reduce human involvement and reduce operating costs. On the other hand, it can improve the network performance and provide better user experience.In this paper, the key technology of SON is deeply studied. A new method of SON self optimization based on double loop structure is proposed and two load balancing algorithms in heterogeneous network scenarios is presented. The main research contents of this paper are as follows:1. A method and apparatus for SON self-optimization based on double loop structure in LTE-A system is proposed. From the macro point of view, the main SON self-optimizing cases are integrated, and the overall framework of SON self-optimizing use cases is constructed. By triggering the decision, the system can adaptively choose the optimized use cases to optimize the system and reduce the conflict of multiple self-optimizing cases simultaneously.2. Based on the Wiener prediction theory, a load balancing algorithm based on load forecasting and user association in LTE-A HetNet system is proposed. By predicting the load state of each node accurately, every node can adjust handover parameters ahead, which can solve the load balancing problem and improve the network performance.(3) A load balancing algorithm based on Q-Learning in LTE-A HetNets is proposed. the reinforcement learning method is applied in the network optimization of load balancing, and see femtocell as a Agent. In each sensing period, agent is constantly listening to the network state, and the Q-Learning algorithm is applied to select a power value as the transmit power, update the Q table, and complete the iteration of the algorithm. Through continuous learning and optimization, the most optimal control strategy is formed gradually. The algorithm can effectively improve the system throughput, and reduce the rate of system blocking rate and dropping rate, and solve the problem of load balancing in heterogeneous networks.