Study Of Cross Layer Optimization Algorithms Based On Ad Hoc Networks

Cross-layer optimization is a new technology which is used in wireless networks for optimizing the structure of networks and the ability of transmission. Different from traditional layered-optimization technologies, it is not an optimization method that is only used in a definite layer which is defined by the OSI structure, but to make the cooperation between the higher network layers and the lower ones to realize the optimization jointly. And the purpose of the optimization is to guarantee the QoS by choosing the optimized strategies of transmission and parameters of protocols.Compared with traditional layered-optimization methods, the main advantage of cross-layer optimization is to make a tradeoff in the qualities of all factors by considering the situation of all layers in the structure of networks, and this kind of optimization method does not just focus on the effect of optimization in a definite layer. Consequently, the requirements of QoS of data transmission in wireless LAN could be better and more efficiently supported by using appropriate cross-layer optimization technologies. In addition, research of cross-layer technologies based on definite networks, such as wireless LAN, Ad hoc networks, and wireless sensor networks is quite meaningful for improving the qualities of networks and strategies design of networks structure.The research in this paper puts emphasis on cross-layer optimization based on Ad hoc networks. After the introduction of Ad hoc networks and main concepts of cross-layer optimization, the methods and basic structure of implementation of cross-layer optimization on Ad hoc networks is described in detail. Then, we analyze the high computational complexity caused by traditional cross-layer optimization strategies selection algorithms, and propose an improved algorithm which performs better in solving computational complexity. And the experiments based on Ad hoc networks show that, compared with traditional optimization methods, the new algorithm could effectively reduce the computational complexity without bringing negative influence to the quality of optimization.