Novel Evolutionary Algorithm for ICA Processor for FPGA Implementation
Jayasanthi Ranjith1, NJR. Muniraj2
1Jayasanthi Ranjith, Electronics and communication Engineering, Anna university,Chennai,Tamilnadu,India.
2NJR.Muniraj,Principal, Tejaa Sakthi Institute of Technology for Women, Coimbatore, TN, India.
Manuscript received on January 01, 2013. | Revised Manuscript received on January 02, 2013. | Manuscript published on January 05, 2013. | PP: 54-57 | Volume-2, Issue-6, January 2013. | Retrieval Number: F1092112612/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Evolutionary programming (EP) has been applied to many numerical and combinatorial optimization problems in recent years. Independent component analysis (ICA) is a statistical signal processing technique for separation of mixed signals, voices and images. The need for evolutionary algorithm for ICA lies in the fact that it needs contrast function optimization which enables the estimation of the independent components. Independent component analysis (ICA) decomposes observed mixed random vectors into statistically independent variables. It aims at finding the underlying independent components in the mixture by searching a linear or nonlinear transformation. It is also more efficient when the cost function, which measures the independence of the components, is optimized. ICA algorithm for contrast function optimization is developed in VHDL .The use of low complexity evolutionary computation with additional operations of mutation and crossover resolves the permutation ambiguity to a large extent. This also ensures the convergence of the algorithm to a global optimum and VLSI implementation results in reduced complexity of algorithms. IEEE single-precision representation, which fits in thirty-two bits, is used for all the manipulations for covering large range of real values.
Keywords: ICA, Evolutionary optimization algorithm, FPGA , Statistical signal processing, VLSI