Research on Blind Equalization Algorithm of Modified RLS Based on Canonical Correlation Analysis
Shunlan Liu, Zhu Fu, Hu Jianhong · 2009
In this paper, a new blind equalization algorithm suitable for single-input multiple-output (SIMO) systems is introduced based on recursive least squares algorithm and canonical correlation analysis (CCA-RLS), and a modified recursive least squares algorithm with variable forgetting factor based on canonical correlation analysis (CCA-MRLS) is proposed. Combining CCA with MRLS, CCA-MRLS algorithm is applied to SIMO systems. Unlike other approaches, the CCA-RLS obtains simultaneously the optimal equalizers as well as the best combination of their outputs. The simulation results demonstrate that the proposed CCA-MRLS algorithm can offer good convergence characteristics and has a faster convergence speed, lower residual inter-symbol interference than CCA-RLS after convergence.