Download now Free registration required
In this paper, adaptive pilot-aided channel estimation for Orthogonal Frequency-Division Multiplexing (OFDM) transmission is addressed. For this purpose, instead of a theoretically optimum Two-Dimensional (2D) filter, two consecutive one-dimensional (2 ? 1D) filters are applied. In particular, three algorithms for the first 1D step are compared, namely Normalized Least-Mean-Squares (NLMS), Normalized Block Least-Mean-Squares (NBLMS) and Recursive Least-Squares (RLS) algorithm. Theoretical analysis of the algorithm performance depending on channel conditions as well as simulation results show that the NBLMS always outperforms the NLMS algorithm, but the RLS only at low carrier-to-noise ratios.
- Format: PDF
- Size: 220.4 KB