Study On Wireless Sensor Network Cross-layer Protocols For QoS Optimization

Wireless sensor network (WSN) is a frontier research field which is formed by a high degree of multi-disciplinary integration; it is one of the best technologies to realize Ambient Intelligence and Artificial Intelligence. In recent years, the networking theory and practical techniques are becoming a study focus with the emergence of many related research achievements. As an important research object of WSN, the QoS performance index system is a parameter set which can evaluate the performance of the networking theory and techniques. To get a comprehensive optimization of QoS index, a thorough study is done on the cross-layer optimization of WSN, the major research contents and productions are shown in the following areas:For the issues of the WSN’s support ability to the requirements of the application techniques, optimization mechanism of network QoS is studied, the QoS’s structure and the relationship between each protocol layer of network and major QoS index are emphatically analyzed. On this basis, new research direction and design task are given for the WSN’s global optimization by proposing a cross-layer optimized design method, which is different from the traditional stratification theory. Moreover, the design methods and solving strategy of the cross-layer model are also studied. All of these works provide the global optimization of WSN with novel research direction and design objective.Regarding the QoS technique requirements to congestion controlling and energy consumption reducing, a new improved algorithm of layered protocol stack is proposed. Research on new MAC protocol algorithm of opportunity adaptive competition mechanism is done. To avoid the data conflict questions, a reasonable backoff window selection strategy is used. Cluster-based routing protocol with high efficiency energy saving is studied; the design method of the optimal detection scope and the node density based algorithm of cluster head selection is used to reduce the network communication energy consumption.To achieve the optimization and jointed control of network QoS performance, the power control based cross-layer optimization design method of WSN is studied. The directional spanning tree routing model is constructed according to the geographical information and transmission power of the nodes; furthermore, study on the optimal intermediate nodes selection mechanism based on the conception of probability distribution energy consumption optimization. To prolong the networks life circle to a maximum extent, optimize the network’s performance, the MAC optimization algorithm based on asynchronous periodicity awake/dormancy mechanism and topology control method based on network loading equilibrium mechanism are studied according to the characteristics of spanning tree.For the technical requirements of the global optimization of network, Network Utility Maximum (NUM) constrained model that reflects global resource and control of network is proposed. Based on the jointed control of transmission rate and power, a Network Utility Maximum based cross-layer protocol is proposed further. According to the theoretic analysis and derivation of network congestion control, the Lagrangian Dual Function based model converting and solving mechanism is discussed to convert multi-objective function to single objective function, and provide more theoretical and technical support to global optimization and schedule of network.In order to resolve the unreliability of wireless link in WSN, the corresponding scenarios of reliable data transmission are analyzed. Based on the cross-layer conception and nodes cooperation mechanism, a node cooperation based multi-path transmission algorithm is proposed. Furthermore, by the selection of optimal intermediate node and a constructed strategy of relaying nodes-cluster, the algorithm enables the end-to-end data transmission of WSN to be more reliable and accurate, and improves the global QoS performance of WSN significantly.Besides study on QoS optimization oriented network cross-layer protocol, algorithm realization and simulation experiment are done in the dissertation. Through analyzing the simulation result, the characteristics and advantages of cross-layer optimized design have been summarized, and the proposed cross-layer optimization mechanisms in the dissertation have been verified to be available and advanced.