Combinations of Multiple Adaptive Lattice Filters in Sparse Cognitive Radio Channel Identification

Abstract

Sparsity awareness feature has been introduced into the a priori recursive least squares (RLS) lattice algorithm by regularizing the RLS cost function with the addition of $\ell_{1}$-norm constraint penalty term. The newly developed sparsity aware algorithm is implemented in the previously proposed lattice filter combination schemes, i.e., Regular Combination of Multiple Lattice Filters (R-CMLF) and Decoupled Combination of Multiple Lattice Filter (D-CMLF) schemes, in cognitive radio (CR) channel identification framework, so that, a s different from other schemes in the literature, sparsity awareness is jointly brought about with the use of different exponential weighting factors in component filters. Hence, fast convergence and low steady state MSD performance are brought together under sparse channel conditions. The performances of lattice component filters with sparsity aware algorithms as well as combination filters of both schemes under white Gaussian input signal conditions are demonstrated by means of mean square deviation (MSD) simulations.