Study On Resource Management Strategies Of The Next Generation Wireless Networks

With the increasing emergence of new-style radio access technologies(RATs) and networking technologies, the wireless networks have been developing tremendously in network coverage, network capacity, transmission rate, service quality, etc. Evolving towards diversified access methods, high-speed data transmission, universal mobility and all IP(Internet Protocol) integration, the next generation wireless networks will ultimately become ubiquitous, all-embracing and omnipotent. The next generation wireless networks integrate different type of mobile networks and Internet, provide smooth access of various wireless technologies and seamless roaming of heterogeneous mobile terminals, and support multiple hops self-organizing, cooperative relay and other new network mode. Connected to the traditional cable networks and Internet by all IP core networks, the next generation wireless networks aim to provide users with ubiquitous Internet access and omnipotent network services by the means of various RATs, flexible networking mode, full network coverage, unprecedented transmission rate and huge network capacity, etc. However, due to the inherent features of wireless networks such as fade, interference, media sharing, user-mobility, and so on, wireless channels are time-varying and capacity limited, which makes the conflict between the growing demand for wireless application and the shortage of wireless resources more and more obvious. The problem mentioned above has become a severe challenge to the development and further innovation of the next generation wireless networks. Therefore, the efficient radio resource management and flexible scheduling strategy, which is one of the main means of solving the contradictions above and also the main research content of our paper, can effectively improve the resource utilization of the existing communication system, expand the available resources space, and improve the overall performance of the wireless communication system.This thesis has mainly studied resource management and scheduling strategy in two typical scenarios of next generation wireless networks(wireless Ad Hoc networks and OFDMA cooperative relay networks), including topology control, cross-layer design, subcarrier selection and pairing, subcarrier power allocation, and so on. The main research contents and contributions of this thesis are as follows:This thesis carefully summarizes the development trends and challenges of wireless networks, and comparatively analyzes the typical communication scenes of next generation wireless networks, including wireless Ad Hoc networks and OFDMA cooperative relay networks. This thesis also particularly investigates the idea and approaches of wireless network resource managements such like topology control, cross-layer design, convex optimization, etc.In this thesis, we study the problem of prolonging network lifetime in large-scale wireless sensor networks where a mobile Sink gathers data periodically along the predefined path. We propose a heuristic topology control algorithm, called Minimum Load Set(MLS) algorithm, which can balance the energy consumption of nodes in the network by appropriately selecting data forwarding path and transmission power for each sensor, so as to extend the life time of the WSN. Furthermore, the thesis discusses how to improve the MLS algorithm to satisfy practical requirements such as distributed computing and transmission timeliness. Theoretical analysis and experimental results show that our algorithm is superior to several earlier algorithms for extending network lifetime.This thesis investigates the cross-layer design in wireless Ad Hoc networks via joint optimization of congestion control at the transport layer, contention control at the medium access control layer(MAC layer), and transmission power control at the physical layer. The goal is to find optimal end-to-end transmission rates at the transport layer, per-link persistence probabilities at the MAC layer and transmitting power at the physical layer to maximize the aggregate source utility. We construct a Generalized Network Utility Maximization(GNUM) problem constrained by multi-dimensional conditions by introducing the auxiliary variables and appropriate variables transform. Then, this thesis proposes the distributed across-layer algorithm based on subgradient. Simulation results further verify the effectiveness and the convergence of our proposed algorithm.Meanwhile, this thesis proposes a joint subcarrier pairing and power allocation(JS2PA) scheme with fairness for Orthogonal Frequency Division Multiple Access(OFDMA) cooperative relay networks. To solve the non-convex JS2 PA problem, firstly, we propose a subcarrier pairing and selection algorithm based on Hungarian method so as to select the appropriate subcarrier pairs for relaying. Secondly, we provide a power allocation algorithm, named PA-MIWD, based on the IWD method in which water drops act as the agents to find the optimal power allocation for each node. Finally, we conduct simulations to validate that the proposed JS2 PA scheme outperforms the existing methods in terms of convergence, total network utility and fairness,and verify the effect of relay location, subcarrier requirement and power allocation on system performance.