Signal recovery via deep convolutional networks
dc.contributor.assignee | Rice University | en_US |
dc.contributor.publisher | United States Patent and Trademark Office | en_US |
dc.creator | Baraniuk, Richard G. | en_US |
dc.creator | Mousavi, Ali | en_US |
dc.date.accessioned | 2021-05-25T20:01:11Z | en_US |
dc.date.available | 2021-05-25T20:01:11Z | en_US |
dc.date.filed | 2017-12-06 | en_US |
dc.date.issued | 2021-04-20 | en_US |
dc.description.abstract | Real-world data may not be sparse in a fixed basis, and current high-performance recovery algorithms are slow to converge, which limits compressive sensing (CS) to either non-real-time applications or scenarios where massive back-end computing is available. Presented herein are embodiments for improving CS by developing a new signal recovery framework that uses a deep convolutional neural network (CNN) to learn the inverse transformation from measurement signals. When trained on a set of representative images, the network learns both a representation for the signals and an inverse map approximating a greedy or convex recovery algorithm. Implementations on real data indicate that some embodiments closely approximate the solution produced by state-of-the-art CS recovery algorithms, yet are hundreds of times faster in run time. | en_US |
dc.digitization.specifications | This patent information was downloaded from the US Patent and Trademark website (http://www.uspto.gov/) as image-PDFs. The PDFs were OCRed for access purposes | en_US |
dc.format.extent | 34 | en_US |
dc.identifier.citation | Baraniuk, Richard G. and Mousavi, Ali, "Signal recovery via deep convolutional networks." Patent US10985777B2. issued 2021-04-20. Retrieved from <a href="https://hdl.handle.net/1911/110644">https://hdl.handle.net/1911/110644</a>. | en_US |
dc.identifier.patentID | US10985777B2 | en_US |
dc.identifier.uri | https://hdl.handle.net/1911/110644 | en_US |
dc.language.iso | eng | en_US |
dc.title | Signal recovery via deep convolutional networks | en_US |
dc.type | Utility patent | en_US |
dc.type.dcmi | Text | en_US |
dc.type.genre | patents | en_US |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- US10985777B2.pdf
- Size:
- 805.4 KB
- Format:
- Adobe Portable Document Format
- Description: