New method improves sparse 2D signal reconstruction with fixed readout budget

This paper introduces a new method for reconstructing sparse 2D signals from subsampled Fourier representations, specifically addressing scenarios where a fixed number of entries can be sampled per channel. Researchers derived a lower bound on the mutual coherence of the compressed sensing matrix for a given readout budget, demonstrating it exceeds the classical Welch bound. The paper also presents deterministic subsampling patterns that achieve this bound for certain matrix dimensions and budgets, with simulations benchmarking their performance against random subsampling. AI

RANK_REASON The item is an academic paper published on arXiv detailing a new method for signal processing. [lever_c_demoted from research: ic=1 ai=0.1]

Read on arXiv cs.IR (Information Retrieval) →

• alphaXiv
• CatalyzeX Code Finder for Papers
• Connected Papers
• CORE Recommender
• DagsHub
• Fourier
• Gotit.pub
• Hugging Face
• Litmaps
• Nitin Jonathan Myers
• ScienceCast
• scite Smart Citations
• Two-dimensional Fourier compressed sensing
• Welch bounds

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →