| With the rapid development of wireless communication technology and the lack of frequency resources, the technology of spectrum sharing becomes more and more important. Cognitive Radio is born to fulfill the needs, which can intelligently and dynamically learn the wireless network environment around and thus an unauthorized system could communicate with each other in taking advantages of spectrum holes of the authorized systems. In a narrow sense of the word, Cognitive Radio is a technology of intelligent spectrum sharing.The spectrum band of 2.3G~2.4GHz is one of the ranges where the 3GPP(The 3rd Generation Partnership Project)LTE(Long Term Evolution) system and some of the radar systems co-exist, but it's utilized inefficiently. From Cognitive Radio, for improving this situation, the radar and LTE systems could share the spectrum band of 2.3G~2.4GHz, and be respectively regarded as authorized and unauthorized systems. So the research work should figure out one of spectrum sensing technologies that LTE system could find the spectrum holes and use it efficiently without interfering the radar systems through proper power control, while performances of LTE system can be optimized.These are the main issues of this research.In the thesis, characteristics of the radar signals are first analyzed. There are kinds of radar signals working in 2.3G~2.4GHz. Uniform impulse signal has been chosen for simulation by Matlab and its characteristics are analyzed for subsequent research preparing. By the comparative analysis of non-cooperative and cooperative detection, we find that these two detection techniques are not suitable to the network environment of this thesis work. So we propose a technology of bi-spectrum slide analysis based on high order cumulant, which could not only automatically suppress the noises, but also effectively detect the spectrum to make it more applicable to the research environment.The unauthorized LTE system can interfere with the authorized radar system and even make it work abnormally, while LTE system is working on spectrum holes of radar system. Proper power control can solve the problem. In this thesis, thes pectrum band of LTE system is close to that of radar sysytem, but not overlapped. The model of coexistent system has been proposed through interference analysis. This thesis has put forward SINR(Signal to Interference plus Noise Ratio) formulas of radar and LTE based on analysis of downlinks and uplinks of LTE system, and then defined utility functions of LTE system. Through power control of downlinks and uplinks, LTE system can be able to correspond and maximize transmitting power. On the condition that radar system is working normally, a improved channel allocation MDCFA(Mixed Distributed/Centralized Control Fair Algorithm) is designed,which is fairer than MDCA(Mixed Distributed/Fair Centralized Algorithm). By combining power control and available channel number matrix, priority matrix PN×K of LTE users is found. Results of simulation by Matlab, the power control scheme proposed is applied to the cognitive network, and MDCFA is more better than MDCA in user access numbers and channel allocation.The innovation points of this thesis are: (1) the spectrum sensing scheme of bi-spectrum slide analysis based on high order cumulant is applied to the given research enviroment,which can detect spectrum holes of radar system; (2) the model of coexistence of radar and LTE systems is put forward, a adaptive power control scheme is proposed, and LTE system works well without interfering radar system; on condition that radar system could work normally, performances of LTE system is optimized by channel allocation. |
PDF Full Text Request