Breaking the coherence barrier: asymptotic incoherence and asymptotic sparsity in compressed sensing
Event details
| Date | 17.04.2013 |
| Hour | 11:00 |
| Speaker |
Anders Hansen, University of Cambridge Bio: Dr Hansen completed his PhD at Cambridge and then moved to Caltech as von Karman Fellow, before returning to Cambridge as a Research Fellow of Homerton College. His research is in computational mathematics and includes applied harmonic analysis, mathematical signal processing with emphasis on sampling theory and compressed sensing as well as spectral theory, computability theory, complexity theory and numerical analysis: a rapidly growing area of mathematics with enormous applications ranging from medical imaging to signal processing. |
| Location | |
| Category | Conferences - Seminars |
In this talk we will discuss new results on how to break the so called coherence barrier in compressed sensing. More precisely, it is well known that compressed sensing relies on two key pillars: incoherence and sparsity. Although there are examples where one has both of these features, there are surprisingly many cases in applications where these criteria are not satisfied, in particular Fourier sampling and wavelet or polynomial reconstruction (both essential in Magnetic Resonance Imaging (MRI)). This has led to a substantial gap between the state of the art compressed sensing theory and its actual use in application. I will show how to bridge this fundamental gap by introducing a new theory where the assumptions of incoherence and sparsity are replaced by two new features: asymptotic incoherence and asymptotic sparsity as well as multi-level sampling. With these new tools at hand I will demonstrate two rather intriguing effects: The success of compressed sensing is resolution dependent and optimal sampling strategies are signal structure dependent.
Practical information
- General public
- Free
Organizer
- Mitra Fatemi <[email protected]>
Contact
- Mitra Fatemi <[email protected]>