Font Size: a A A

Research On Power Control In TD-SCDMA System Based On Robust RBF Neural Networks

Posted on:2010-03-12Degree:MasterType:Thesis
Country:ChinaCandidate:C MaFull Text:PDF
GTID:2248330395458102Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
In recent years, along with the rapid development of the mobile communication systems, the2G mobile communication systems lagged far behind the current people’s high-quality expectation of networks. Therefore, the3G mobile communication system is emerged. The3G mobile communication system is the new generation system which satisfies with the International Mobile Telecommunications-2000(IMT-200). TD-SCDMA (Time Division-Synchronous Code Division Multi Access)system, which is the3G mobile communication system researched by China, provides the basis voice services and low to high data rate services. It also supports the multimedia services such as voice and video telephony, high-speed internet access, etc. Consequently, TD-SCDMA system has attracted more attention due to its advantages.In the course of transferring information, there are several disadvantages such as time-selective fading, frequency-selective fading, near-far effect, multiple access interference (MAI), white Gaussian noise, etc, which affect the reliability of TD-SCDMA system and lower the capacity of TD-SCDMA system. Power control is one effective way to avoid the near-far problem and to increase the capacity in TD-CDMA system. Due to forward-link is more essential, we only considered the forward-link close-loop power control in TD-SCDMA system throughout this paper.The main work and contributions in the paper are as follows:the principle of robust RBF neural networks based on robust learning algorithm is investigated deeply. Then its performance is illustrated by two examples. The simulation results have shown that the method enhances the network robustness against outliers. Therefore, study is conducted on the application of robust RBF neural networks to the forward-link closed-loop power control. We design the contracture of controller based on robust RBF neural networks. The practical performance of this controller is simulated via computer. The simulation results have shown that the controller can avoid the outliers affect and obtain the satisfactory effectiveness. Finally, the research work involved in the thesis is summarized and the future developments in power control about TD-SCDMA are forecasted.
Keywords/Search Tags:3G mobile communication system, TD-SCDMA mobile communication, forward-link, outliers, robust learning algorithm, RBF neural networks
PDF Full Text Request
Related items