An Efficient VLSI Architecture for FastICA by Using the Algebraic Jacobi Method for EVD

Blind source separation (BSS) is a problem that appears in many research fields. Fast Independent components analysis (FastICA) one of the techniques to solve problem. The researchers have verified effectiveness technique through offline public datasets. development real-time portable system involving such computationally complex requires an efficient hardware implementation FastICA. A Field programmable gate array (FPGA) and application-specific integrated circuit (ASIC) are two promising platforms implement This work proposes new method, called ALgebraic Jacobi Method (ALJM), for performing eigenvalue decomposition (EVD) required We use simplification, polynomial approximation, Newton-Raphson method calculating rotation. In this way, we ensure reusability between EVD stage weight vector estimation (WVE) FastICA which reduces computational complexity power consumption, without compromising its computation speed. evaluate ALJM-based by BSS on linear mixtures deterministic random signals comparing performance results with existing methods. After verifying functionality numerical stability, propose scalable systolic processing (SPA) it Spartan-6 FPGA. By implementations FastICA, terms speed, area, power, conclude most methods prototyping commercializing comprising