On complex-valued deautoconvolution of compactly supported functions with sparse Fourier representation

Convergence rates results for the Tikhonov regularization of nonlinear ill-posed operator equations are missing, even for a Hilbert space setting, if a range type source condition fails and if moreover nonlinearity conditions of tangential cone type cannot be shown. This situation applies for a deautoconvolution problem in complex-valued L 2-spaces over finite real intervals, occurring in a slightly generalized version in laser optics. For this problem we show that the lack of applicable convergence rates results can be overcome under the assumption that the solution of the operator equation has a sparse Fourier representation. Precisely, we derive a variational source condition for that case, which implies a convergence rate immediately. The surprising observation is that a sparsity assumption imposed on the solution leads to success, although the used norm square is not known to be a sparsity promoting penalty in the Tikhonov functional.

[1]  Ronny Ramlau,et al.  CONVERGENCE RATES FOR REGULARIZATION WITH SPARSITY CONSTRAINTS , 2010 .

[2]  Bernd Hofmann,et al.  About a deficit in low-order convergence rates on the example of autoconvolution , 2015 .

[3]  O. Scherzer,et al.  A convergence rates result for Tikhonov regularization in Banach spaces with non-smooth operators , 2007 .

[4]  Barbara Kaltenbacher,et al.  Regularization Methods in Banach Spaces , 2012, Radon Series on Computational and Applied Mathematics.

[5]  B. Hofmann,et al.  Phase retrieval via regularization in self-diffraction based spectral interferometry , 2014, 1412.2965.

[6]  J. Coyle.  Inverse Problems , 2004 .

[7]  Regularization of an autoconvolution problem occurring in measurements of ultra-short laser pulses , 2011 .

[8]  D. Lorenz,et al.  Convergence rates and source conditions for Tikhonov regularization with sparsity constraints , 2008, 0801.1774.

[9]  Jens Flemming,et al.  Generalized Tikhonov regularization , 2011 .

[10]  Bernd Hofmann,et al.  An extension of the variational inequality approach for obtaining convergence rates in regularization of nonlinear ill-posed problems , 2010 .

[11]  Bernd Hofmann,et al.  Parameter choice in Banach space regularization under variational inequalities , 2012 .

[12]  Martin Burger,et al.  Convergence rates in $\mathbf{\ell^1}$-regularization if the sparsity assumption fails , 2012 .

[13]  Bernd Hofmann,et al.  Regularization of an autoconvolution problem in ultrashort laser pulse characterization , 2013, 1301.6061.

[14]  Otmar Scherzer,et al.  Variational Methods in Imaging , 2008, Applied mathematical sciences.

[15]  Bernd Hofmann,et al.  On autoconvolution and regularization , 1994 .

[16]  M. Grasmair.  Generalized Bregman distances and convergence rates for non-convex regularization methods , 2010 .

[17]  Ronny Ramlau,et al.  TIGRA—an iterative algorithm for regularizing nonlinear ill-posed problems , 2003 .

[18]  H. Engl,et al.  Convergence rates for Tikhonov regularisation of non-linear ill-posed problems , 1989 .

[19]  H. Engl,et al.  Regularization of Inverse Problems , 1996 .

[20]  P. Maass,et al.  Sparsity regularization for parameter identification problems , 2012 .

[21]  Bernd Hofmann,et al.  Variational regularization of complex deautoconvolution and phase retrieval in ultrashort laser pulse characterization , 2016 .

[22]  Bernd Hofmann,et al.  On inversion rates for the autoconvolution equation , 1996 .

[23]  Bernd Hofmann,et al.  Approximate source conditions for nonlinear ill-posed problems—chances and limitations , 2009 .

[24]  Bernd Hofmann,et al.  Regularization properties of the sequential discrepancy principle for Tikhonov regularization in Banach spaces , 2014 .