System identification
Fundamentals of statistical signal processing: estimation theory
Fundamentals of statistical signal processing: estimation theory
Independent component analysis, a new concept?
Signal Processing - Special issue on higher order statistics
Partially blind estimation: ML-based approaches and Cramer-Rao bound
Signal Processing
An analytical constant modulus algorithm
IEEE Transactions on Signal Processing
Estimating evoked dipole responses in unknown spatially correlatednoise with EEG/MEG arrays
IEEE Transactions on Signal Processing
IEEE Transactions on Signal Processing
Fast maximum likelihood for blind identification of multiple FIRchannels
IEEE Transactions on Signal Processing
Strict identifiability of multiple FIR channels driven by anunknown arbitrary sequence
IEEE Transactions on Signal Processing
Blind channel and carrier frequency offset estimation usingperiodic modulation precoders
IEEE Transactions on Signal Processing
Bounds on bearing and symbol estimation with side information
IEEE Transactions on Signal Processing
Parameter estimation problems with singular information matrices
IEEE Transactions on Signal Processing
Exploring estimator bias-variance tradeoffs using the uniform CRbound
IEEE Transactions on Signal Processing
IEEE Transactions on Information Theory
Hi-index | 0.00 |
Constrained Cramér-Rao bounds are developed for convolutive multi-input multi-output (MIMO) channel and source estimation in additive Gaussian noise. Properties of the MIMO Fisher information matrix (FIM) are studied, and we develop the maximum rank of the unconstrained FIM and provide necessary conditions for the FIM to achieve full rank. Equality constraints on channel and signal parameters provide a means to study the potential value of side information, such as training symbols (semi-blind case), constant modulus (CM) sources, or known channels. Nonredundant constraints may be combined in an arbitrary fashion, so that side information may be different for different sources. The bounds are useful for evaluating the performance of SIMO and MIMO channel estimation and equalization algorithms. We present examples using the constant modulus blind equalization algorithm. The constrained bounds are also useful for evaluating the relative value of different types of side information, and we present examples comparing semi-blind, constant modulus, and known channel constraints. While the examples presented are primarily in the communications context, the CRB framework applies generally to convolutive source separation problems.