Robust Channel Estimation Technique for OFDM Systems

Abstract

Orthogonal Frequency Division Multiplexing (OFDM) is a multi-carrier transmi-ssion technique in wireless environments, and can be seen as a multi-carrier digital modulation or multi-carrier digital multiplexing one as well. A large number of orthogonal sub-carriers are used to transmit information. OFDM systems have high utilization of the frequency spectrum and satisfactory capabilities of reducing multi-path inference. So, OFDM has been considered as one of the core techno-logies of 4th generation (4G) wireless communication systems.

Channel estimation plays a very important role in OFDM systems. It can generally be separated into two methods: pilot-based channel estimation and blind channel estimation. Pilot-based channel estimation, which is the focus of this thesis, estimates the channel information by obtaining the impulse response from all sub-carriers by pilot. Compared to blind channel estimation, which uses statistical information of the received signals and is not considered in this thesis, pilot-based channel estimation is a practical and effective method.

This thesis covers the basic principles of the OFDM system, system construction and the advantages and disadvantages of OFDM systems are considered also. It also offers a brief overview on signal propagation, channel parameters and the basic principles of channel estimation in OFDM systems.

The great challenge of channel estimation methods is to compromise between low complexity and high performance. In this thesis three improved methods of channel estimation are introduced. These methods are based on pilot-aided OFDM system with the arrangement employed in the DVB-T2 standard in time-varying frequency-selective fading channels. The first and second proposed methods (low complexity and improved low complexity methods, respectively) are modified methods based on the Domain Transform Least Square Estimation (DTLSE) method; they reduce the computational complexity by avoiding the use of the matrix inversion. The estimation matrix size for obtaining Channel Impulse Response (CIR) depends only on the length of the channel rather than the number of pilot sub-carriers or the size of OFDM symbols.