Abstract:
Channel estimation plays a crucial role in wireless communication receivers with coherent detection. In contrast to training-based methods, the blind (self-recovering) approach to channel estimation in which the estimate is done purely based on the knowledge of channel output is addressed in this work. Currently in practice, most of the wireless standards employ training-based methods or pilot subcarriers for estimation of CIR. Blind techniques are being researched on from past three decades. The advantage of adopting blind techniques is the conservation of signal bandwidth through the elimination of training/ pilot symbols. This transforms into higher spectral efficiency and thus higher information rates can be achieved at given channel bandwidth.
Recently, orthogonal frequency division multiplexing (OFDM) has become an attractive choice in wireless standards. Further, use of multiple antennas at both ends of a wireless link: multiple-input multiple-output (MIMO) technology has been demonstrated to have the potential of achieving extraordinary data rates. The use of MIMO technology in combination with OFDM, i.e., MIMO-OFDM, therefore seems to be an attractive solution for future broadband wireless systems.
Blind channel estimation using second-order statistics in SISO-OFDM and MIMO-OFDM systems are addressed. To be more specific, the two most sought out approaches namely: subspace decomposition method and precoder-induced-correlation averaging method are described and compared in terms of their performance and practical applicability. Finally, the techniques to resolve the constant complex scalar estimation ambiguity fundamental to all second-order statistics based methods are addressed. A novel completely/ totally blind channel estimation technique via source constellation-splitting and modified phase-directed algorithm for SISO-OFDM systems is proposed and evaluated.
it
