Article

Partial-Update L∞ -Norm Based Algorithms

Dept. of Electr. & Comput. Eng., Concordia Univ., Montreal, Que.
Circuits and Systems I: Regular Papers, IEEE Transactions on (Impact Factor: 2.3). 03/2007; DOI: 10.1109/TCSI.2006.883863
Source: IEEE Xplore

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

0 Bookmarks
 · 
75 Views
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes an affine projection algorithm (APA) with selective input vectors, which based on the concept of selective-update in order to reduce estimation errors and computations. The algorithm consists of two procedures: input- vector-selection and state-decision. The input-vector-selection procedure determines the number of input vectors by checking with mean square error (MSE) whether the input vectors have enough information for update. The state-decision procedure determines the current state of the adaptive filter by using the state-decision criterion. As the adaptive filter is in transient state, the algorithm updates the filter coefficients with the selected input vectors. On the other hand, as soon as the adaptive filter reaches the steady state, the update procedure is not performed. Through these two procedures, the proposed algorithm achieves small steady-state estimation errors, low computational complexity and low update complexity for colored input signals.
    Proceedings of SPIE - The International Society for Optical Engineering 10/2011; DOI:10.1117/12.913391 · 0.20 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes a non-periodic-partial-update affine projection algorithm with data-selective updating. The proposed algorithm employs two update concepts: non-periodic partial update and data-selective update. The former plays a role in adjusting the length of the update period, and the latter in reducing computational complexity. Thus, the algorithm requires two key procedures of length assignment and state decision. The length assignment procedure determines the length of the update period by checking whether the current input vectors have enough information with an update period assignment criterion. The state decision procedure stochastically determines whether the adaptive filter has reached a steady state. When the current state of the adaptive filter is confirmed as a transient state by the decision procedure, the algorithm updates all filter coefficients with an update period assigned by the length assignment procedure. Through these two procedures, the proposed algorithm not only achieves good performance, especially for colored input signals, in terms of the convergence rate and steady-state estimation errors but also provides a substantial reduction in the number of updates.
    Intelligent Signal Processing and Communications Systems (ISPACS), 2013 International Symposium on; 01/2013
  • [Show abstract] [Hide abstract]
    ABSTRACT: Two robust affine projection sign (RAPS) algorithms, both of which minimize the mixed norm of $l_1$ and $l_2$ of the error signal, are proposed. The direction vector of the RAPS algorithms is obtained from the gradient of an $l_1$ norm-based objective function, while two related $l_2$ norm-based minimization problems are solved to obtain the line search of the two RAPS algorithms. The $l_1$ norm-based direction vector reduces the impact of impulsive noise, whereas the $l_2$ norm-based line search produces an unbiased solution in the proposed algorithms. In addition, one of the two RAPS algorithms shares the data selective adaptation used in the set-membership (SM) affine projection (SMAP) algorithm. The proposed algorithms are shown to offer a significant improvement in the convergence speed as well as a significant reduction in the steady-state misalignment relative to the pseudo affine projection sign (PAPS) algorithm. In addition, the proposed algorithms offer robust performance with respect to impulsive noise and improved tracking of the unknown system in comparison to that provided by the PAPS and Affine projection sign (APS) algorithms. These features of the proposed algorithms are demonstrated using simulation results in system-identification and echo-cancellation applications.
    IEEE Transactions on Signal Processing 01/2014; 62(13):3349-3359. DOI:10.1109/TSP.2014.2324997 · 3.20 Impact Factor