A VLSI design of singular value decomposition processor for portable continuous-wave diffusion optical tomography systems

Abstract

In recent years, developments in diffuse optical tomography (DOT) technology have rapidly progressed in biomedical applications such as functional brain imaging. Furthermore, during the last decade, singular value decomposition has increasingly found many applications in the engineering field. Our goal is to develop a truly portable DOT system by utilizing system-on-chip design technique. In our DOT system, designing the SVD processor using VLSI technique is very efficient for solving the inverse problem in the DOT algorithm. In this paper, we focus on the hardware architecture design of the Jacobi singular value decomposition (JSVD) for applications in continuous wave diffusion optical tomography (CW-DOT) systems. The design is implemented using 90nm CMOS process technology and simulation results verify the functionality of the JSVD design within the developed CW-DOT system.