Reproduce “Neural Network-Assisted CLEAN for Efficient Multipath Parameter Estimation in MIMO Systems” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Chaofan Deng, Linyu Sun, Jaeho Lee, Zishen Wan, Arijit Raychowdhury
NN-CLEAN replaces repeated angle-of-arrival, angle-of-departure, and distance searches in CLEAN channel estimation with a neural classifier, while retaining least-squares gain fitting and physical reconstruction with path subtraction. At batch size eight, it completes three-path extraction about 27× faster than grid-search CLEAN, with 52-fold lower peak GPU memory. On synthetic tests, grid-search CLEAN is more accurate for three and four paths.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.