Research On Rate And Power Optimization For Wireless Multi-hop Cognitive Networks

With the rapid development of mobile communication and Internet technology, the demand of spectrum is increasing day by day. As an effective way to alleviate the problem of spectrum scarcity, cognitive radio technology has attracted much attention. Wireless multi-hop cognitive network is composed of nodes with cognitive capabilities and owns the characteristic of the multi-hop structure, which allows nodes to connect with each other through wireless links. How to perform optimal allocation for network resources has become a research hotspot. In the thesis, it takes the wireless multi-hop cognitive network as background, by means of cross-layer optimization, explores optimization of data rate and power consumption.Firstly, to maximize data rate in wireless multi-hop cognitive network, it applies the physical interference model and combines the signal to interference plus noise ratio(SINR) in the physical layer, scheduling in the link layer and routing in the network layer as constraints, which formulates a mixed integer nonlinear program(MINLP) problem. The method of Reformulation Linearization Technique(RLT) is applied to relax the nonlinear constraint into the linear one to obtain one optimal solution, which construct an upper bound of the optimal solution to the MINLP problem. Then, to the MINLP problem, a feasible solution based on the upper bound is regarded as the lower bound. Simulation results demonstrate the efficacy of the solution procedure and obtain SINR, power for each node at time slots and date rate on each link.Secondly, to perform optimal power control in wireless multi-hop cognitive network, it considers the protocol interference model, analyzes necessary and sufficient condition of successful transmission for nodes and combines power control in the physical layer, scheduling in the link layer and routing in the network layer as constraints, which formulate a mixed integer nonlinear program(MINLP) problem. To solve it, we develop a piece-wise linearization technique to transform the nonlinear term in constraints into the linear one and finally obtain the approximate optimal solution. Simulation results demonstrate the efficacy of the solution procedure and optimal power control can achieve energy saving.In summary, the optimization of rate and power on wireless multi-hop cognitive shows important theoretical significance and practical value.