how to optimize some algorithm in cuda.

Spiking Neural Networks in C++ with strong GPU acceleration through CUDA

parallel algorithm based on cuda

Shared memory overlap-and-save method for NVIDIA GPUs using CUDA

usrp stream program, single channel dsp with CUDA, single channel dsp without CUDA (for comparison)

Grating is a GPU-based polyphase filter bank (PFB) spectrometer.

PFB code implementation in CUDA C++