Diffusion Estimation Algorithm for Mixtures of Factor Analyzers with Common Factor Loadings in Sensor Network

In this paper, we propose a diffusion estimation algorithm based on mixtures of factor analyzers with common factor loadings (MCFA) in sensor network. For each node, its objective function is defined as the linear combination of log-likelihoods about its neighborhood. Then, the diffusion estimation algorithm for the MCFA is derived in details. Furthermore, convergence of the proposed algorithm is analyzed. Finally, the performance of this algorithm is evaluated by an application example. Experimental results validate the promising performance of the proposed algorithm.