| Grant number: | 14/22258-2 |
| Support type: | Scholarships abroad - Research Internship - Doctorate |
| Effective date (Start): | March 01, 2015 |
| Effective date (End): | July 31, 2015 |
| Field of knowledge: | Physical Sciences and Mathematics - Physics |
| Principal Investigator: | Renato Vicente |
| Grantee: | Paulo Victor Camargo Rossi |
| Supervisor abroad: | Yoshiyuki Kabashima |
| Home Institution: | Instituto de Física (IF). Universidade de São Paulo (USP). São Paulo , SP, Brazil |
| Local de pesquisa : | Tokyo Institute of Technology, Yokohama (Tokyo Tech), Japan |
| Associated to the scholarship: | 14/00792-7 - Statistical mechanics applied to compressed sensing, BP.DR |
Abstract
Compressed Sensing (CS) é a Family of techniques where signal measurement and signal compression happen simultaneously. Given a signal represented by a N-dimensional vector x0 and a measuring matrix F of dimension MxN, we make M linear measurements y=F x0. Previous work by Sompolinsky et al. and Kabashima et al. where Statistical Mechanics techniques were used indicated that when x0 is k-sparse (i.e., when it has at most k non-zero entries), there is a regime where linear programming algorithms can reconstruct the original signal x0 from the knowledge of F and y even when M | |