Abstract : A semi closed-form expression of the Fisher information matrix in the context of K-distributed observations with parameterized mean is given and related to the classical, i.e. Gaussian case. This connection is done via a simple multiplicative factor, which only depends on the intrinsic parameters of the texture and the size of the observation vector. Finally, numerical simulation is provided to corroborate the theoretical analysis
Mohammed Nabil El Korso, Alexandre Renaux, Philippe Forster. CRLB under K-distributed observation with parameterized mean. IEEE Sensor Array Multichannel Workshop, Jun 2014, A Coruña, Spain. ⟨hal-00993010⟩
