JISE


  [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]


Journal of Information Science and Engineering, Vol. 33 No. 1, pp. 51-62


Hardware Implementation of Local Mean Decomposition


PEI-YIN CHEN, YEN-CHEN LAI AND PING-HSUAN LAI
Department of Computer Science and Information Engineering
National Cheng Kung University
Tainan, 701 Taiwan
E-mail: {pychen; P78001324; P76014101}@mail.ncku.edu.tw


    Local mean decomposition (LMD) is an effective signal analysis method for analyzing nonlinear and nonstationary signals. LMD has been usefully applied in a wide variety of applications. However, achieving real-time LMD calculations in software is difficult. In this paper, a flexible, low-cost, and high-performance hardware architecture for LMD is proposed that satisfies the real-time requirements of various LMD applications. All proposed circuits were developed using Verilog and then synthesized using the Synopsys Design Compiler with the Taiwan Semiconductor Manufacturing Company 0.18-μm cell library. With the help of parameterization, the proposed LMD circuit can easily be used for various applications and hardware architectures.


Keywords: hardware, local mean decomposition (LMD), real-time, signal processing, de-modulation

  Retrieve PDF document (JISE_201701_04.pdf)