Journals: Parameter estimation in wireless sensor networks with faulty transducers: a distributed EM approach
S. Silva Pereira, R. López-Valcarce and A. Pagès Zamora


We address the problem of distributed estimation of a vector-valued parameter performed by a wireless sensor network in the presence of noisy observations which may be unreliable due to faulty transducers. The proposed distributed estimator is based on the Expectation-Maximization (EM) algorithm and combines consensus and di usion techniques: a term for information di usion is gradually turned o , while a term for updated information averaging is turned on so that all nodes in the network approach the same value of the estimate. The proposed method requires only local exchanges of information among network nodes and, in contrast with previous approaches, it does not assume knowledge of the a priori probability of transducer failures or the noise variance. A convergence analysis is provided, showing that the convergent points of the centralized EM iteration are locally asymptotically convergent points of the proposed distributed scheme. Numerical examples show that the distributed algorithm asymptotically attains the performance of the centralized EM method.

©UPC Universitat Politècnica de Catalunya
Signal Processing and Communications group
Powered by Joomla!.