Multiplierless Realization of Linear DSP Transforms by Using Common Two-Term Expressions
Journal of VLSI Signal Processing Systems
Hi-index | 35.68 |
In this correspondence, statistical methods are used to estimate the error on the output signal due to quantization of filter coefficients and the fractional part of the data bus in FIR-based multirate systems. In addition, a convex programming model is formed to estimate the filter coefficient wordlengths given the output signal error. The model is capable of handling folded architectures as well as regular ones