This is an implementation of the CLEAN algorithm which identifies individual waveforms in a multi-component signal by simple spectral summations. The details of the algorithm can be found in …

We introduce the one dimensional FFT algorithm in this section, which will be used in our GPU implementation. The DFT of a vector of size N can be rewritten as a sum of two smaller DFTs, …

Mar 27, 2017 · EP- and PSO-based CLEAN have a higher computational burden, because they must solve a three-dimensional optimization problem using the stochastic search method. To …

high-performance parallel computers with large number of processing elements (cores). This paper introduces a popular software package called P3DFFT implementing Fast F. urier …

Fast Fourier Transforms perform a vital role in many applications from astronomy to cellphones. The complexity of these algorithms results from the many computational steps, including …

The common technique is to grid the sampled data points and evaluate the discrete Fourier transform via an FFT. We grid using a convolution with a standard spheroidal function and use …

It implements 3D FFT and related algorithms such as Chebyshev transform (an important class of algorithm for simulations in a wide range of fields). P3DFFT uses 2D, or pencil, decomposition. …

Support for fast parallel complex-to-complex and real-to-complex transforms of 3-D FFTs and 4-D FFTs with arbitrary input size. Tuning of communication with an auto-tuning feature. Support …

An implementation of the CLEAN algorithm for extracting individual wave packets from a multi-component signal - jhsa26/clean_algorithm

In this technical report, we design mixed-precision and approximate 3D FFTs that trade-off speed for accuracy. We target hybrid architectures with GPUs.