https://hgpu.org/?p=12239
Improvement Study of EEMD Decomposition Efficiency Based on CUDA Architecture