CudaFilters: A SignalPlant library for GPU‐accelerated FFT and FIR filtering. (18th May 2017)

• Record Type: Journal Article
• Authors: Nejedly, Petr; Plesinger, Filip; Halamek, Josef; Jurak, Pavel
• Is Part Of: Software, practice & experience. Volume 48:Number 1(2018)
• Page Start: 3
• Page End: 9
• Publication Date: 2017-05-18
• Subjects: CUDA -- FFT filter -- FIR filter -- GPU acceleration -- SignalPlant; Computer software -- Periodicals; Computer programming -- Periodicals; Computer programs -- Periodicals; 005.3
• DOI: 10.1002/spe.2507
• Languages: English
• ISSNs: 0038-0644

Abstract

Summary: Signal filtering is one of the essential tasks in signal processing. It may become an extremely time‐consuming process, as in the case of intracranial electroencephalogram recordings (eg, 30‐min records) with a large number of channels (up to 256) and high sampling frequencies (up to 5 kHz in research related to ultra‐high‐frequency oscillations). The usual way of dealing with time consumption is process parallelization. Moreover, parallelization using graphic processing unit (GPU) allows further shortening of computing times thanks to the large number of GPU cores. This paper describes a library for GPU‐accelerated finite impulse response (FIR) and fast Fourier transform (FFT) filtering—"CudaFilters." This library is designed for SignalPlant software—a free tool for signal analysis. The resultant acceleration in computing times was 5× to 40× depending on the task, data, and hardware configuration. The results were also compared to computing speeds in Matlab.