Partial-Update L∞ -Norm Based Algorithms
ABSTRACT The computational complexity of an adaptive filtering algorithm increases with increasing the filter tap length and therefore, the use of such a filter can become prohibitive for certain applications, especially for real-time implementation. In this paper, we develop low-complexity adaptive filtering algorithms by incorporating the concept of partial updating of the filter coefficients into the technique of finding the gradient vector in the hyperplane based on the Linfin-norm criterion. Two specific partial update algorithms based on the sequential and M-Max coefficient updating are proposed. The statistical analyses of the two algorithms are carried out, and evolution equations for the mean and mean-square of the filter coefficient misalignment as well as the stability bounds on the step size are obtained. It is shown that the proposed partial update algorithm employing the M-Max coefficient updating can achieve a convergence rate that is closest to that of the full update algorithm. Finally, simulations are carried out to validate the theoretical results and study the convergence rate of the proposed algorithms
- SourceAvailable from: ieice.org[Show abstract] [Hide abstract]
ABSTRACT: This paper provides a partial-update normalized sign least-mean square (NSLMS) algorithm with sparse up-dates. The proposed algorithm reduces the computational complexity compared with the conventional L ∞ -norm adap-tive filtering algorithms by decreasing the frequency of up-dating the filter coefficients and updating only a part of the filter coefficients. And we develop a mean square analysis to present the convergence of the proposed algorithm. Experi-mental results show that the proposed algorithm has the good convergence performance with greatly reduced computational complexity.
- [Show abstract] [Hide abstract]
ABSTRACT: We present a novel affine projection algorithm (APA) which automatically determines its projection order by an evolutionary method. The evolutionary method increases or decreases the projection order by comparing the output error with a threshold. The experimental results show that the proposed algorithm has fast convergence speed and small steady-state error compared to the conventional APA.IEEE Signal Processing Letters 12/2009; · 1.67 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Subband adaptive filters (SAFs) have been widely used due to their fast convergence rate for colored input signals. In order to meet the conflicting requirements of fast convergence rate and small misalignment of SAFs, a variable step-size matrix (VSSM) has been presented for the recently proposed normalized SAF (NSAF). The VSSM is derived by letting the powers of the subband a posteriori errors of the NSAF equal the powers of the corresponding subband system noises at each time instant and therefore the subband system noise powers should be estimated. In this paper, we propose to incorporate the concept of periodic update into the subband system noise power estimates to reduce the computational complexity of the VSSM-NSAF. Simulation results in a system identification context show that the proposed periodic update based method can achieve a good performance close to that of the full update one.01/2010;