Repozytorium publikacji - Politechnika Gdańska

The generalized improved version of the alpha max plus beta min square-rooting algorithm and its realization in the Field Programmable Gate Array (FPGA) are presented. The algorithm computes the square root to calculate the approximate magnitude of a complex sample. It is especially useful for pipelined calculations in the DSP. In case of four approximation regions it is possible to reduce the peak error form 3.95% to 0.33%. This is attained by determination of the approximate ratio of arguments and adequate selection of algorithm coefficients. Four approximation regions are used and hence four sets of coefficients. Also a Xilinx FPGA implementation for 12-bit sign magnitude numbers is shown.

Autorzy

• dr inż. Robert Smyk link otwiera się w nowej karcie ,
• dr hab. inż. Maciej Czyżak link otwiera się w nowej karcie