Power Allocation For System Optimization In Dense Small Cell Networks

Small cell network is one of the key technologies to meet the soaring re-quirement of the fifth generation mobile system. In small cell networks, pico cells, home eNodeBs and relays are all called small cells. Small cells can be equipped with not only the same frequency band as the macro cell, but also a higher frequency band that is not suitable for macro cell. The scenario of dense deployed small cells is the major feature of small cell networks, providing ad-ditional coverage of macro cells and extremely high data traffic. In dense small cell networks, the channel gains are improved since the distance between eN-odeBs and users are decreased a lot. However, the strong interference among small cells becomes the bottle neck of the improvement of system performance. In this paper, power allocation is adopted to optimize the quality of service, sys-tem throughput and system energy efficiency in dense small cell networks.The following aspects are covered in this thesis:In the research of quality of service, this thesis studies the system with limited resources, targeting at maximizing the number of small cells which can satisfy the traffic requirement. The system model that each small cell serves single user is formulated and a distributed algorithm based on game theory is designed. Moreover, the victim small cells are compensated further in the re-sults of Nash equilibrium of the designed game.In the research of system throughput, we target at the system throughput maximization in the small cell networks. Firstly, the system model that each small cell serves multi users is developed. Secondly, a novel signaling called Inter-cell-Signal to Interference plus Noise Ratio (ISINR) is defined and mod-ified to show the algebraic properties of the system capacity. With the help of ISINR, we have an easy way to identify the local monotonicity of the system capacity. Thirdly, on each subchannel in each iteration, we divide the small cell eNodeBs into different subsets and the optimal power allocation strategies are designed.In the research of network energy efficiency, the target is to maximize the data transmitted by every Joule. With the small cell model with multi users, the energy efficient maximization problem is formulated. Since the fractional problem is difficult to solve, we transfer the problem into an equivalent iterative optimization problem which has a subtractive form. In this case, the problem can be solved by a distributed algorithm with low computational complexity.