A Nonlinear Filtering Method for Geometric Subspace Tracking
نویسنده
چکیده
We formulate the problem of tracking principal subspaces as a problem in nonlinear ltering. The subspaces are represented by their complex projection-matrices, and moving subspaces correspond to trajectories on the Grassmann man-ifold. Taking a Bayesian approach, we impose a smoothness prior on the subspace rotation. Combining ideas from importance sampling and sequential methods, we apply a recursive Monte Carlo approach to solving for MMSE estimates.
منابع مشابه
Geometric Filtering for Subspace Tracking
We address the problem of tracking principal subspaces using ideas from nonlinear ltering. The subspaces are represented by their complex projection-matrices, and time-varying subspaces correspond to trajectories on the Grassmann manifold. Under a Bayesian approach, we impose a smooth prior on the velocities associated with the subspace motion. This prior combined with any standard likelihood f...
متن کاملNon-linear Anti-Geometric Diffusion based Adaptive Singular Value Decomposition Method for Efficient Image De-noising
˗Image de-noising is an important challenging issue in image pre-processing. Two popular methods to the problem are partial differential equation (PDE) based nonlinear diffusion method and singular value decomposition (SVD) method. Various image de-noising algorithms based on these two methods have been independently developed. This paper proposes an approach for image de-noising by performing ...
متن کاملBayesian Nonlinear Filtering via Information Geometric Optimization
In this paper, Bayesian nonlinear filtering is considered from the viewpoint of information geometry and a novel filtering method is proposed based on information geometric optimization. Under the Bayesian filtering framework, we derive a relationship between the nonlinear characteristics of filtering and the metric tensor of the corresponding statistical manifold. Bayesian joint distributions ...
متن کاملThe Kalman Filtering Blind Adaptive Multi-user Detector Based on Tracking Algorithm of Signal Subspace
Multi-user detection is an effective method to reduce multiple access interference in code division multiple access (CDMA) systems. This paper discusses a signal subspace based blind adaptive multiuser detector and a Kalman filtering blind adaptive multiuser detector. Combining them together, a new Kalman filtering blind adaptive multiuser detector based on a tracking algorithm of the signal su...
متن کاملAdaptive Neural Network Method for Consensus Tracking of High-Order Mimo Nonlinear Multi-Agent Systems
This paper is concerned with the consensus tracking problem of high order MIMO nonlinear multi-agent systems. The agents must follow a leader node in presence of unknown dynamics and uncertain external disturbances. The communication network topology of agents is assumed to be a fixed undirected graph. A distributed adaptive control method is proposed to solve the consensus problem utilizing re...
متن کامل