85
reads

Lih-Ren Lin

2010-03-26
09:00:00 - 11:00:00

201 , Old Mathematics Building

The restricted isometry property (RIP) is a concept proposed by Candes and Tao to measure, roughly speaking, how closely does a matrix behave like an isometry. Such concept, as Donoho and Huo’s “incoherence”, is invented to give criterions in the theory of compressed sensing. More precisely, we can show that if a matrix A enjoys good RIP, then a sparse signal x can be recovered from the measurement Ax by a convex optimization problem. The adjective “compressed” comes from the fact that the dimension of y may be much smaller than that of x. In this lecture, we will briefly review the idea of compressed sensing, the concept of “incoherence” and “restricted isometry”. Then we’ll focus on questions and developments related to RIP.