Constrained adaptive LMS L-filters

نویسندگان

  • Constantine Kotropoulos
  • Ioannis Pitas
چکیده

Two novel adaptive nonlinear filter structures are proposed which are based on linear combinations of order statistics. These adaptive schemes are modifications of the standard LMS algorithm and have the ability to incorporate constraints imposed on coefficients in order to permit location-invariant and unbiased estimation of a constant signal in the presence of additive white noise. The convergence in the mean and in the mean square of the proposed adaptive nonlinear filters is studied. The rate of convergence is also considered. It is verified by simulations that the independence theory provides useful bounds on the rate of convergence. The extreme eigenvalues of the matrix which controls the performance of the location-invariant adaptive LMS L-filter are related to the extreme eigenvalues of the correlation matrix of the ordered noise samples which controls the performance of other adaptive LMS L-filters proposed elsewhere. The proposed filters can adapt well to a variety of noise probability distributions ranging from the short-tailed ones (e.g. uniform distribution) to long-tailed ones (e.g. Laplacian distribution). Zusammenfassung. Es werden zwei neue adaptive nichtlineare Filterstrukturen vorgeschlagen, die auf Linearkombinationen yon Order-Statistik beruhen. Diese adaptiven Strukturen sind Modifikationen des iiblichen LMS-Algorithmus und erlauben die Einbringung yon Bedingungen bezfiglich der Koeffizienten, um ortsinvariante und erwartungstreue Schfitzungen konstanter Signale unter additivem, weiBem Rauschen zu erm6glichen. Die Konvergenz bez/iglich des Mittelwertes und des quadratischen Mittelwertes wird fiir die vorgeschlagenen nichtlinearen Filter untersucht. Weiterhin wird die Konvergenzgeschwindigkeit betrachtet. Durch Simulationen wird gezeigt, dab die Independence-Theorie brauchbare Grenzen ffir die Konvergenzrate liefert. Die extremen Eigenwerte der Matrix, die das Verhalten des ortsinvarianten adaptiven LMS L-Filters bestimmt, werden den Eigenwerten der Korrelationsmatrix der geordneten Rauschabtastwerte gegeniibergestellt, die das Verhalten anderer adaptiver LMS L-Filter bestimmt. Die vorgeschlagenen Filter stellen sich sehr gut auf eine Vielzahl verschiedener Verteilungsdichten des Rauschens ein, angefangen von schmalen Verteilungen (z.B. Gleichverteilung) bis hin zu langsam abfallenden (z.B. Laplace). R6sum+. Nous proposons deux structures de filtre non-lin~aire originales, structures bastes sur des combinaisons lin6aires de statistiques d'ordre. Ces techniques adaptatives sont des modifications de I'algorithme LMS standard et ont la capacit6 d'incorporer des contraintes impos+es sur les coefficients afin de permettre une estimation ne variant pas selon la localisation et non biais6e d 'un signal constant en pr6sence de bruit blanc additif. Nous 6tudions la convergence en moyenne et en moyenne quadratique des filtres non-lin6aires adaptatifs propos6s. Nous consid~rons 6galement le taux de convergence. Nous v+rifions par des simulations que l'hypoth+se d'ind6pendance fournit des bornes utiles sur le taux de convergence. Nous relions les valeurs propres extremes de la matrice qui contr61e les performances du L-filtre LMS adaptatif ne variant pas selon la localisation aux valeurs propres extremes de la matrice de correlation des 6chantillons de bruit ordonn+s qui contr61e les performances d'autres L-filtres LMS proposes ailleurs. Les filtres propos6s peuvent s 'adapter ais+ment 5. une vari6te de distributions de densit6 de bruit allant de celles 5. queue courte (p.e. la distribution uniforme) 5. celles 5. queue longue (p.e. la distribution de Laplace).

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Adaptive LMS L-Filters for Smoothing Noisy Images

Several adaptive LMS L-filters, both constrained and unconstrained ones, are developed for noise suppression in images and being compared in this paper. First, the location-invariant LMS L-filter for a nonconstant signal corrupted by zero-mean additive white noise is derived. Subsequently, the normalized and the sign LMS L-filters are studied. It is shown that both these filters turn to be iden...

متن کامل

Speech Enhancement by Modified Convex Combination of Fractional Adaptive Filtering

This paper presents new adaptive filtering techniques used in speech enhancement system. Adaptive filtering schemes are subjected to different trade-offs regarding their steady-state misadjustment, speed of convergence, and tracking performance. Fractional Least-Mean-Square (FLMS) is a new adaptive algorithm which has better performance than the conventional LMS algorithm. Normalization of LMS ...

متن کامل

Analysis of Adaptive Filters in Feedback Cancellation for Sinusoidal Signals

Analysis of coupling effects in adaptive filters is presented for feedback cancellation systems in hearing aids. As the incoming signal, a sum of sinusoids in white noise is assumed in this paper. In the proposed scheme two LMS-type adaptive filters are used. The one is set for identifying the impulse response of the feedback path. The other is associated with frequency estimation. We propose a...

متن کامل

Mean-Square Performance of the Constrained LMS Algorithm

—The so-called constrained least mean-square algorithm is one of the most commonly used linear-equality-constrained adaptive filtering algorithms. Its main advantages are adaptability and relative simplicity. In order to gain theoretical insights into the performance of this algorithm, we examine its mean-square convergence and derive an expression for its steady-state mean-square deviation. Ou...

متن کامل

Cellular LMS L-filters for Noise Suppression in Still Images and Image Sequences

A novel class of nonlinear adaptive L-filters based on cellular neural networks topology is presented. Like cellular neural systems and cellular automata as well, processing nodes, called cells, communicate with each other directly only through its nearest neighbors exchanging information. Each cell is an adaptive LMS L-filter. The proposed filters share the best features of both adaptive filte...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • Signal Processing

دوره 26  شماره 

صفحات  -

تاریخ انتشار 1992