Reserach On Offloading Scheme In Framless Network Architecture

In recent years, with the rapid development of mobile Internet services, the data traffic demand for terminal applications is growing. Then how to improve the mobile communication system capacity on the basis of current mobile communications network architecture by the evolution of network architecture or networking technology has become a hot issue, so the research on network architecture and networking technology is imperative. By opening the traditional base station equipment, Frameless Network Architecture (FNA) make the network upgrade and expansion more simple, more suited to the current versatile wireless network environment. By leveraging the complementary characteristics of heterogeneous network, offloading technology not only achieves the system capacity improvement without changing the current network architecture, but also can be applied to FNA. The main contents of this dissertation include:To solve the serious interference and low utilization of part of base stations problem for high-density heterogeneous hybrid network scenarios, we propose FNA, which splits the traditional base station into the antenna element being responsible for signal transmission and reception and centralized processing element being responsible for signal processing. Moreover, we introduce the concept of resource pooling, which achieves the centralized resource management, the separation of control plane and data plane. After integrating Software Defined Network (SDN) and OpenFlow technology into FNA, we use the Flow Table to implement the resource management and routing and make the functional modules in the network architecture reconfigurable. To solve the uneven load distribution and energy inefficiency problem in high-density heterogeneous hybrid network scenarios, we propose auction-based energy efficiency optimized offloading scheme. The scheme adopts both the forward auction and reverse auction strategies to optimize the system energy efficiency under the constrain of the transmission rate of user equipment. The auction result is to select the appropriate offloading target cell, while allocating the power and bandwidth to the user equipment. The simulation results show that the proposed offloading scheme can provide higher system energy efficiency and more offloading gain, compared with the comparative algorithms.Facing the situation of various service and base station type coexistence problem in high-density heterogeneous hybrid network scenarios, we propose the adaptive offloading scheme based on fuzzy logic. At first, based on the result of neural network based traffic load distribution prediction algorithm, the system allocates the resources between heterogeneous networks. Considering both the power efficiency and matching degree, the offloading decision is made to select the most appropriate offloading target cell. The simulation results show that the traffic load distribution prediction scheme brings performance gains for the offloading scheme.To solve uneven load distribution problem in time domain of heterogeneous network, we propose the delay offloading scheme, in which the total consumption of the network is defined. For the network consumption, the occupation of macrocell network and the user satisfaction is taken into account at the same time. Besides, the penalty function is defined to compensate for the loss of user satisfaction caused by delay. When the penalty function is a monotone convex function, we can find a threshold structure to simplify the computing complexity of the delay offloading scheme. The simulation results show that the delay offloading scheme can reduce overall network consumption.