Blind separation of sources, part II: Problems statement
نویسندگان
چکیده
Though it arouses more and more curiosity, the HJ iterative algorithm has never been derived in mathematical terms to date. We attempt in this paper to describe it from a statistical point of view. For instance the updat ing term of the synaptic efficacies matrix cannot be the gradient of a single C a functional contrary to what is sometimes understood. In fact, we show that the HJ algorithm is actually searching common zeros of n functionals by pipelined stochastic iterations. Based on simulat ion results, advantages and limitations as well as possible improvements are pointed out after a short theoretical analysis. Zusammenfassung. Obwohl er mehr und mehr Interesse weckt, ist der iterative HJ-algori thmus noch nie au f mathemat ischem Wege hergeleitet worden. In diesem Beitrag versuchen wir, ihn von einem statistischen Gesichtspunkt aus zu beschreiben. Zum Beispiel kann der Erneuerungsterm der synaptischen Wirksamkeitsmatrix nicht der Gradient eines einzelnen C 2Funktionals sein obwohl er gelegentlich so interpretiert wird. Wir zeigen, dab der HJ-Algori thmus in der Tat gemeinsame Nullstellen von n Funkt ionalen mittels stochastischer, im Pipelining ausgefiihrter Iterationen sucht. Auf der Grundlage von Simulat ionen werden Vorteile und Bergrenzungen sowie mrgl iche Verbesserungen nach kurzer theoretischer Analyse dargelegt. R6sum6. Bien qu'il 6veille maintenant de plus en plus de curiosit6, l 'algorithme it&atif HJ n 'a jusqu '~ pr6sent jamais 6t6 justifi6 en termes math6mat iques rigoureux. Par exemple, le terme de remise ~ jour des poids synaptiques ne peut pas ;~tre le gradient d 'une fonctionnelle unique de classe C 2, contrairement h ce qui est paffois reconnu. Nous essayons dans cet article de le pr6senter en empruntant une approche statisique. Nous montrons que I 'algorithme HJ recherche en r6alit6 les z6ros communs de n fonctionnelles par des it6rations stochastiques pipelin6es. Apr~s une courte analyse th6orique, et en nous appuyant sur des simulations, nous soulignons les avantages et les limitations de l 'algorithme, ainsi qu 'un certain nombre d'am61iorations possibles.
منابع مشابه
Blind Separation of Jointly Stationary Correlated Sources
The separation of unobserved sources from mixed observed data is a fundamental signal processing problem. Most of the proposed techniques for solving this problem rely on independence or at least uncorrelation assumption for source signals. This paper introduces a technique for cases that source signals are correlated with each other. The method uses Wold decomposition principle for extracting ...
متن کاملResearch of Blind Signals Separation with Genetic Algorithm and Particle Swarm Optimization Based on Mutual Information
Blind source separation technique separates mixed signals blindly without any information on the mixing system. In this paper, we have used two evolutionary algorithms, namely, genetic algorithm and particle swarm optimization for blind source separation. In these techniques a novel fitness function that is based on the mutual information and high order statistics is proposed. In order to evalu...
متن کاملIntelligent location of simultaneously active acoustic emission sources: Part II
Part I describes an intelligent acoustic emission locator, while Part II discusses blind source separation, time delay estimation and location of two continuous acoustic emission sources. Acoustic emission (AE) analysis is used for characterization and location of developing defects in materials. AE sources often generate a mixture of various statistically independent signals. A difficult probl...
متن کاملResearch of Blind Signals Separation with Genetic Algorithm and Particle Swarm Optimization Based on Mutual Information
Blind source separation technique separates mixed signals blindly without any information on the mixing system. In this paper, we have used two evolutionary algorithms, namely, genetic algorithm and particle swarm optimization for blind source separation. In these techniques a novel fitness function that is based on the mutual information and high order statistics is proposed. In order to evalu...
متن کاملExtraction of Sensory part of Ulnar Nerve Signal Using Blind Source Separation Method
A recorded nerve signal via an electrode is composed of many evokes or action potentials, (originated from individual axons) which may be considered as different initial sources. Recovering these primitive sources in its turn may lead us to the anatomic originations of a nerve signal which will give us outstanding foresights in neural rehabilitations. Accordingly, clinical interests may be r...
متن کامل¨ Ozgür Yılmaz Part I: Overview of Research Program
Research Statement – September 2, 2008 In this statement, I summarize my contributions in the areas of applied and computational harmonic analysis, signal processing, and information theory. In particular, my work focuses on (i) approximation theory for quantization of redundant expansions, (ii) blind source separation, and (iii) sparse approximations. Part I gives a high level summary of the g...
متن کاملذخیره 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 ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Signal Processing
دوره 24 شماره
صفحات -
تاریخ انتشار 1991