Implementation And Optimization Of Frequency Domain Equalization Algorithm Based On Single Carrier

One of the most challenging problems in high data rate wireless communication is to overcome the time dispersion caused by multipath propagation. OFDM is a popular modulation scheme for broadband wireless communication. It elegantly handles multipath and has low complexity. Recently, single-carrier with frequency-domain processing (SC-FDE) is gaining attention as a possible competitor for OFDM. It is based on the same principles and, therefore, it has similar multipath handling capabilities and low complexity. Moreover, SC-FDE promises several other attractive advantages, such as power amplifier efficiency and inherent frequency diversity exploitation. Frequency domain equalization algorithms based on single carrier are studied in this thesis.This thesis first analyses the characterestics of multipath fading in the wireless channel and WiMAX channel model. According to the standard IEEE 802.16, Stanford University Interity (SUI) channel model is employed in our simulation platform. Different channel effects can be simulated by different type of SUI channels.Then, the principles of OFDM and SC-FDE are analyzed and compared. Based on SUI channel model, linear frequency domain equalization, including Zero Forcing equalization (ZF) and Minimal Mean Square Equalization (MMSE), and nonlinear frequency domain equalization, including Decision Feedback Equalization (FD-DFE), are simulated under SC-FDE simulation platform. The results show that nonlinear frequency domain equalization get better performance than the linear frequency domain equalization at the expense of increased complexity.Next, decision feedback equalization based on noise prediction (FDE-NP), one of nonlinear frequency domain equalizations, is investigated in depth. In particular, based on FDE-NP, three new algorithms are proposed and simulated, that is, (1) inverse feedback algorithm; (2) Selective feedback algorithm;(3) Optimal selective feedback algorithm. Our results show that the inverse feedback algorithm can reduce the decision delay but retain the same error performance as the traditional algorithm; the selective feedback algorithm can get better error performance with the sacrifice in complexity; the optimal selective feedback algorithm can achieve better performance at the expense of increased complexity.