Research On Self-Optimization Technologies Of Ofdma-Based Femtocell Networks

With rapid growth of indoor mobile traffic in recent years, improving indoor radio coverage has recently become more and more important for voice, video and high data-rate services. The orthogonal frequency division multiple access (OFDMA)-based Femtocell network is assumed a good solution to improve indoor coverage and system capacity. And the Self-organizing Network (SON) is the key technology for Femtocell to reduce the costs, ease of use and optimize the overall network performance, including self-configuration, self-optimization and self-healing.The research work is sponsored by China National "863" Program:"Wireless communication system employing spectrum resource sharing"(Grant No.2009AA011802); Intergovernmental cooperation project with the University of London:"Semi-intelligent energy-saving cellular network based on interference management". The self-optimization technologies of Femtocell network are researched in this paper, including interference management, energy-efficiency optimization and coverage optimization, so as to realize the automatic, intelligent and adaptive Femtocell network. The main research work and achievements of this thesis are as follows:(1) The related theory and technologies of self-optimization for Femtocell network are analyzed. Firstly, the characteristics of the existing indoor coverage technologies are analyzed, including repeater, distributed antenna systems, radiating cable, Picocell and Femtocell. Then, the research development and standardization process of Femtocell network are summarized. And the main research questions for self-optimization of Femtocell network are also pointed out.(2) The interference management method based on swarm intelligence optimization theory and power allocation is proposed to deal with the interference issue in OFDMA Femtocell network. Firstly, the objective of power allocation is modeled with different utility functions for macrocell and Femtocell, respectively. Then the power allocation algorithms with Particle Swarm Optimization (PSO) and Differential Evolution (DE) are proposed, respectively. Simulation results show that both the PSO and DE are albe to quickly and efficiently solve the power allocation problem. And the PSO has a faster convergence speed, while the DE has higher convergence precision. Meanwhile, the proposed power allocation algorithm can effectively improve the outage probability and throughput performance of macrocell users with minor impact on Femtocell users.(3) An energy-efficient distributed power control algorithm is proposed to optimize the energy efficiency of OFDMA Femtocell network. The proposed algorithm takes into account the constraints of subchannel maximum transmit power and the minimum target signal to interference plus noise ratio (SINR), so as to maximize the energy efficiency. And the power control problem is solved by augmented Lagrange multiplier method. The proposed power control algorithm can effectively optimize the energy efficiency of the OFDMA Femtocell network with the minimum target SINR constraint, even in arge scale deployment scenarios.(4) The effective capacity-based parameter optimization algorithm of Femtocell sleep mode is proposed to optimize the energy efficiency of OFDMA Femtocell network. The definition of effective capacity is extended to formulate the tradeoff relationship between energy efficiency and waiting delay of users. Then two parameter optimization algorithms of sleep mode are proposed:Maximize energy efficiency with effective capacity constraint; and Maximize effective capacity with energy efficiency constraint. Simulation results show that the proposed algorithm can effectively optimize the energy efficiency of OFDMA Femtocell network, and obtain the optimal tradeoff between energy efficiency and user waiting delay. (5) A coverage optimization algorithm based on the Multiple Attribute Decision Making (MADM) is proposed to optimize the radio coverage area of OFDMA Femtocell network in enterprise deployment. Considering some basic requirements of coverage, including coverage holes, coverage overlaps, area spectral efficiency (ASE) and load balancing, the coverage optimization problem is modeled as a MADM problem. Then the MADM-based coverage optimization problem is solved by the Gray relation analysis (GRA) method. Simulation results show that the proposed scheme can achieve a better ASE performance and effectively balance the traffic load amongst all Femtocells, as well as obtaining a good tradeoff between coverage holes and coverage overlap. Furthermore, the results of sensitive analysis for the MADM-based coverage optimization are also described.