Identifiability in Bilinear Inverse Problems With Applications to Subspace or Sparsity-Constrained Blind Gain and Phase Calibration

Bilinear inverse problems (BIPs), the resolution of two vectors given their image under a bilinear mapping, arise in many applications. Without further constraints, BIPs are usually ill-posed. In practice, the properties of natural signals are exploited to solve BIPs. For example, subspace constraints or sparsity constraints are imposed to reduce the search space. These approaches have shown some success in practice. However, there are few results on uniqueness in BIPs. For most BIPs, the fundamental question of under what condition the problem admits a unique solution is yet to be answered. For example, blind gain and phase calibration (BGPC) is a structured BIP, which arises in many applications, including inverse rendering in computational relighting (albedo estimation with unknown lighting), blind phase and gain calibration in sensor array processing, and multichannel blind deconvolution (MBD). It is interesting to study the uniqueness of such problems. In this paper, we define identifiability of a BIP up to a group of transformations. We derive necessary and sufficient conditions for such identifiability, i.e., the conditions under which the solutions can be uniquely determined up to the transformation group. These conditions take the form of dividing the identifiability of the pair of unknown variables into the individual identifiability of each variable. Although verifying these individual conditions requires problem-specific procedures, this framework is universally applicable to all BIPs. Applying these results to BGPC, we derive sufficient conditions for unique recovery under several scenarios, including subspace, joint sparsity, and sparsity models. For BGPC with joint sparsity or sparsity constraints, we develop a procedure to compute the transformation groups corresponding to inherent ambiguities. We also give necessary conditions in the form of tight lower bounds on sample complexities, and demonstrate the tightness of these bounds by numerical experiments. The results for BGPC not only demonstrate the application of the proposed general framework for identifiability analysis, but are also of interest in their own right.

[1]  Sanjeev Arora,et al.  New Algorithms for Learning Incoherent and Overcomplete Dictionaries , 2013, COLT.

[2]  Xiaodong Li,et al.  Solving Quadratic Equations via PhaseLift When There Are About as Many Equations as Unknowns , 2012, Found. Comput. Math..

[3]  Praneeth Netrapalli,et al.  J ul 2 01 4 A Clustering Approach to Learn Sparsely-Used Overcomplete Dictionaries , 2014 .

[4]  Deepa Kundur,et al.  Blind image deconvolution , 1996, IEEE Signal Process. Mag..

[5]  J R Fienup,et al.  Phase retrieval algorithms: a comparison. , 1982, Applied optics.

[6]  Aditya Bhaskara,et al.  More Algorithms for Provable Dictionary Learning , 2014, ArXiv.

[7]  Minh N. Do,et al.  MCA: A Multichannel Approach to SAR Autofocus , 2009, IEEE Transactions on Image Processing.

[8]  Prateek Jain,et al.  Learning Sparsely Used Overcomplete Dictionaries via Alternating Minimization , 2013, SIAM J. Optim..

[9]  Thomas Kailath,et al.  Direction of arrival estimation by eigenstructure methods with unknown sensor gain and phase , 1985, ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[10]  Karim Abed-Meraim,et al.  Blind system identification , 1997, Proc. IEEE.

[11]  Yonina C. Eldar,et al.  Phase Retrieval via Matrix Completion , 2011, SIAM Rev..

[12]  L. Taylor.  The phase retrieval problem , 1981 .

[13]  Minh N. Do,et al.  Subspace methods for computational relighting , 2013, Electronic Imaging.

[14]  Lang Tong,et al.  A new approach to blind identification and equalization of multipath channels , 1991, [1991] Conference Record of the Twenty-Fifth Asilomar Conference on Signals, Systems & Computers.

[15]  Jonathan H. Manton,et al.  On the algebraic identifiability of finite impulse response channels driven by linearly precoded signals , 2005, Syst. Control. Lett..

[16]  Sunav Choudhary,et al.  Sparse blind deconvolution: What cannot be done , 2014, 2014 IEEE International Symposium on Information Theory.

[17]  Yanjun Li,et al.  Identifiability in Blind Deconvolution With Subspace or Sparsity Constraints , 2015, IEEE Transactions on Information Theory.

[18]  Emmanuel J. Candès,et al.  PhaseLift: Exact and Stable Signal Recovery from Magnitude Measurements via Convex Programming , 2011, ArXiv.

[19]  Yoram Bresler,et al.  Uniqueness in bilinear inverse problems with applications to subspace and joint sparsity models , 2015, 2015 International Conference on Sampling Theory and Applications (SampTA).

[20]  G. E. Bredon.  Introduction to compact transformation groups , 1972 .

[21]  Eric Moulines,et al.  Subspace methods for the blind identification of multichannel FIR filters , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[22]  Justin K. Romberg,et al.  Blind Deconvolution Using Convex Programming , 2012, IEEE Transactions on Information Theory.

[23]  Chrysostomos L. Nikias,et al.  EVAM: an eigenvector-based algorithm for multichannel blind deconvolution of input colored signals , 1995, IEEE Trans. Signal Process..

[24]  Michael Elad,et al.  Dictionaries for Sparse Representation Modeling , 2010, Proceedings of the IEEE.

[25]  Sunav Choudhary,et al.  Identifiability Scaling Laws in Bilinear Inverse Problems , 2014, ArXiv.

[26]  Hui Liu,et al.  Recent developments in blind channel equalization: From cyclostationarity to subspaces , 1996, Signal Process..

[27]  Praneeth Netrapalli,et al.  A Clustering Approach to Learning Sparsely Used Overcomplete Dictionaries , 2013, IEEE Transactions on Information Theory.

[28]  Karim Abed-Meraim,et al.  Robustness of blind subspace based techniques using ℓp quasi-norms , 2010, 2010 IEEE 11th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC).

[29]  Huan Wang,et al.  Exact Recovery of Sparsely-Used Dictionaries , 2012, COLT.

[30]  Michael Elad,et al.  Optimally sparse representation in general (nonorthogonal) dictionaries via ℓ1 minimization , 2003, Proceedings of the National Academy of Sciences of the United States of America.

[31]  Antonio M. Peinado,et al.  Diagonalizing properties of the discrete cosine transforms , 1995, IEEE Trans. Signal Process..