Research On Intelligent Control And Optimization Strategy In Self-Organizing Networks

The accelerated integration of mobile communication and internet, the rapid development of wireless access technology and the sustainable growth in the number of intelligent terminal and types of high bandwidth service, make the future wireless mobile communication network under tremendous pressure in terms of network capacity, user experience, energy efficiency, and etc. In addition, the higher requirement of network performance will increase the network capital expenditure (CAPEX) and operating Expenditure (OPEX). How to utilize the limited network resource more efficiently and improve the cost benefit of operators has become an important research topic in the area of wireless communication.To reduce the cost of network construction and operation, to improve the utilization of network resource and to guarantee the quality of network operation, self-organizing network (SON) emerged as required. Self-organizing network aims to enhance the network self-organization ability and realize the self-organization function in each functional area. Self-organizing network technology has been widely concerned by many standardization organizations and research institutions. However, with the sustained development and evolution of future mobile communication technology, the network environment will become more complex and new network characteristics and requirements will emerge. Therefore, it is necessary to enhance the network self-organization ability, improve the network performance, and promote the further research and application of self-organizing network.This paper mainly studies the coverage and capacity optimization, energy saving of base station and cell outage compensation problems in self-organizing network. The main contributions of this paper include the following aspects.In the research of coverage and capacity optimization, the factors which affect coverage and capacity optimization are analyzed and the optimization problem is modeled. To solve the modeled coverage and capacity optimization problem, a hybrid coverage and capacity optimization architecture is proposed. In addition, in view of that the self-organization ability of future wireless communication network should be able to adapt to the system load fluctuation in time and space and allocate the networks resource according to the needs, a coverage and capacity optimization scheme using fuzzy neural network based on reinforcement learning (RL-FNN) is proposed. All SON entities cooperatively adapt their antenna tilt and power configurations using fuzzy neural network based on reinforcement learning. And a central mechanism of sharing optimization experience enables distributed cooperation-based learning. The simulation results show that RL-FNN is able to acquire robust optimization policies for different complex scenarios and maintains a significantly better performance in terms of coverage and capacity with low energy consumption. This especially results in a dramatic improvement in energy efficiency.In the research of energy saving of base station, an active/sleep scheduling scheme based on user activity sensing of small cells is proposed in K-tier heterogeneous networks. The scheme enhances the user activity sensing ability of base stations which can decide the active/sleep status accordingly. Coverage probability, network capacity, and energy consumption of the proposed scheme in K-tier heterogeneous networks are analyzed using stochastic geometry, accounting for cell association uncertainties due to random positioning of users and BSs, channel conditions, and interference. In addition, a fuzzy Q learning-based sensing probability optimization (SPO) approach is proposed to optimize the user activity sensing probability of each small cell tier, considering user activity fluctuations and user QoS. Simulation results showed that SPO achieves low energy consumption with guaranteed network capacity and coverage probability.In the research of cell outage compensation, to utilize the network resources efficiently and provide service compensation to users within the cell outage area, a coalition game based cooperative resource allocation algorithm is proposed based on the modeling and analysis of cell outage compensation problem. In the proposed algorithm, small cells play coalition games to form a set of coalitions which determines the allocation of subchannels. And each coalition of small cells serves a user cooperatively with optimized power allocation. Simulation results show that the proposed algorithm can solve network failure problem effectively, and both cell capacity and user fairness are improved at the same time especially when lots of small cells are faulty.