Research On Particle Swarm Optimization Algorithm And Its Applications In Cross Layer Optimization

With the wireless communication networks rapid development and the growing demand of business, contradiction between the limited radio resources and continuously improvement the quality of service is becoming more and more sharp. Cross-layer optimization problem is the key theoretical innovation of the next-generation wireless communication system. The method is to break the idea of the traditional hierarchical optimization. Take the original segmentation network layers as a unified to design, optimization and control, and make full use of the relevant information between the layers. The overall optimization of wireless network protocols achieves the purpose of efficient utilization of wireless resourcesThis thesis is mainly to solve the cross-layer optimization problems in multi-user OFDM system and the WSN system. Compared with the traditional solution there is marked improvement in solving the cross-layer optimization problem in multi-user OFDM system and WSN system. This thesis first introduces relevant theoretical knowledge, and describes the mechanism of particle swarm optimization and cross-layer optimization in detail. And then the OFDM system and how to allocate resource in multi-user OFDM systems dynamic are described. Communication problems are real engineering problems with constraints, complexity and modeling difficulties. The most commonly methods to solve such problems is intelligent optimization algorithms. Compared with the classical optimization algorithm, the intelligent optimization algorithm is simple, fast convergence, highly parallel, self-organized and strong robustness. In this thesis, the particle swarm optimization algorithm is introduced for its fast convergence and less parameters needed. For the different QoS requirements in multi-user OFDM system, a dynamic resource allocation scheme is used. User scheduling priority is to consider the CSI and the queue status information of the MAC layer. We dynamically allocate power to maximize the system capacity with the particle swarm optimization algorithm after finishing the subcarrier allocation Wireless sensor networks is a special self-organizing network with large number of nodes, limited computing ability and battery power. The system is more and more widely applied in the present and the associated optimization problems naturally becomes the research hotspot. In the thesis the network lifetime of the wireless sensor networks is studied to solve the limited nodes’ battery power without timely replacement. We hope to obtain better energy efficiency extending the lifetime of the entire network.LEACH stands for Low Energy Adaptive Clustering Hierarchy for low-power wireless sensor network routing design widely being used for its better energy efficiency and simple procedures but with the shortcoming of the. In the end of the thesis, we improve the LEACH protocol to solve the problem of the cluster node excessive burden. A new function is constructed, which takes the node residual energy and distance nodes into account. Experimental results show that the improvement algorithm is a great improvement comparing to the LEACH protocol. Meanwhile, we also conduct a theoretical study of multi-sensor information fusion.In this thesis, the main line is the specific application of the particle swarm optimization algorithm, and algorithm analysis and engineering applications are fully integrated. The highlight is the research work of the particle swarm optimization algorithm in cross-layer resource allocation for the wireless communication having a good guide for increasingly scarce communication resources. In this paper, the mutual convergence between the chapters has good continuity.