Cognitive Radio, Adaptive Transmit Signal Design Study

To solve the contradiction between explosively increasing demand of divert wireless service and the scarcity of spectrum resource, Cognitive Radio (CR) was proposed as a highly promising solution. In CR systems, CR Users (CRU) can recognize the spectrum hole both in time and space, which are the spectrum bands not occupied by the Licensed Users (LU, also called Primary users, PU), and transmit their data on these bands. Cognitive Radio has great impact on improving the spectrum efficiency.This thesis focus on the problem of how to design the transmit signal of Cognitive Radio systems. The main concerns are to optimize the usage of the resource of Cognitive radio system and, in the meantime, to protect the transmission of Licensed Users from being damaged.The aspects investigated in this thesis are as following:For single user multicarrier based Cognitive Radio system, how to suppress the high energy leakage of the OFDM symbol is a significant issue since this kind of leakage could bring harm to the transmission of Licensed Users. A concise review of some main existing methods and analysis of the incompatibility of the special requirement of CR systems are given in this part firstly. Besides that, two improved formulations and solution methods which can guarantee the protection level of the Licensed Users and the suboptimal capacity of CRUs are proposed.Moreover, for multiuser CR cells, in which the CR base station is with multiple antennas and the CR users are with single one, this thesis formulate the problem about avoiding interference to the Licensed Users and guaranteeing the receive SINR of the CRUs as a downlink joint power control and beamforming problem. By mathematical operations, this thesis demonstrates this problem is a standard convex optimization problem and thus can be solved by Second Order Cone Programming (SOCP). For the practical reason such as to realize the algorithm by hardware, this thesis prove that the SOCP problem can be solved as an uplink problem by fixed point iteration method via Lagrange duality. The simulation results show that the performance of fixed point method can well approximate the optimal one.Besides, this thesis also studies the uplink problem under the scenario of CR users and CR base station both with multiple antennas in CR cells. A pre-filtering procedure and a multiuser detection procedure are put in the transmit end and the receiver, respectively, to assist the CR users to avoid the interference to each other. The interference temperature is introduced to estimate the interference level to the LUs. Subgradient method is used for solving the formulated problems and the simulations show that the transmit and receive structure proposed is better than others.