In this article, the problems of decentralized Singular Value Decomposition (d-SVD) and decentralized Principal Component Analysis (d-PCA) are studied, which are fundamental in various signal processing applications. Two scenarios of d-SVD are considered depending on the availability of the data matrix under consideration. In the first scenario, the matrix of interest is row-wisely available in each local node in the network. In the second scenario, the matrix of interest implicitly forms an outer product generated from two different series of measurements. Combining the lightweight local rational function approximation approach and parallel averaging consensus algorithms, two d-SVD algorithms are proposed to cope with the two aforementioned scenarios. We demonstrate the proposed algorithms with two respective application examples for Extremely Large-scale Antenna Array (ELAA) systems: decentralized sensor localization via low-rank matrix completion and decentralized passive radar detection. Moreover, a non-trivial truncation technique, which employs a representative vector that is orthonormal to the principal signal subspace, is proposed to further reduce the associated communication cost with the d-SVD algorithms. Simulation results show that the proposed d-SVD algorithms converge to the centralized solution with reduced communication cost compared to those facilitated with the state-of-the-art decentralized power method.