Title :
GPU based implementation of a 64-channel polyphase channalizer
Author :
Amean Al-safi;Bradley Bazuin
Author_Institution :
Electrical and Computer Engineering, Western Michigan University, Kalamazoo, Michigan, USA
Abstract :
Polyphase filter bank channalizers are essential components in numerous communication systems involving signal separation and sample rate conversion. In this paper, a novel fully parallelized 64 channels polyphase channalizer has been implemented using general purposes graphic processing units (GPGPU). The implemented models have been compared with a corresponding serial implemented model. The comparison indicates that the parallel models provide a computation speed of 9-16 times faster than the serial one using Nvidia GeForce GT520M and Quadro 600 GPUs.
Keywords :
"Graphics processing units","Filter banks","Discrete Fourier transforms","Computational modeling","MATLAB","Kernel"
Conference_Titel :
Circuits and Systems Conference (DCAS), 2015 IEEE Dallas
DOI :
10.1109/DCAS.2015.7356605
