Showing papers by "Mats Viberg published in 2006"

On the asymptotic performance analysis of subspace DOA estimation in the presence of modeling errors: case of MUSIC

Abstract: This paper provides a new analytic expression of the bias and RMS error (root mean square) error of the estimated direction of arrival (DOA) in the presence of modeling errors. In , first-order approximations of the RMS error are derived, which are accurate for small enough perturbations. However, the previously available expressions are not able to capture the behavior of the estimation algorithm into the threshold region. In order to fill this gap, we provide a second-order performance analysis, which is valid in a larger interval of modeling errors. To this end, it is shown that the DOA estimation error for each signal source can be expressed as a ratio of Hermitian forms, with a stochastic vector containing the modeling error. Then, an analytic expression for the moments of such a Hermitian forms ratio is provided. Finally, a closed-form expression for the performance (bias and RMS error) is derived. Simulation results indicate that the new result is accurate into the region where the algorithm breaks down.

Calibrating an Array with Scan Dependent Errors Using a Sparse Grid

Abstract: Today most arrays have a very high mechanical and electrical accuracy Our aim is to allow for larger errors (low manufacturing cost), while keeping the calibration grid sparse (low calibration cost) To be able to handle the imperfections causing scan dependent errors (like position errors), we suggest a new array receive calibration method based on local models The results show that our method performs much better than linear interpolation or global (direction independent) calibration regarding direction of arrival estimation using MUSIC for arrays with large position errors In beamforming for eg communication, however, the global calibration performs better (lower side lobes) since it is optimized for the whole calibration region, while the local calibration is optimized for one direction

Enhanced spatial-range mean shift color image segmentation by using convergence frequency and position

Abstract: Mean shift is robust for image segmentation through local mode seeking. However, like most segmentation schemes it suffers from over-segmentation due to the lack of semantic information. This paper proposes an enhanced spatial-range mean shift segmentation approach, where over-segmented regions are reduced by exploiting the positions and frequencies at which mean shift filters converge. Based on our observation that edges are related to spatial positions with low mean shift convergence frequencies, merging of over-segmented regions can be guided away from the perceptually important image edges. Simulations have been performed and results have shown that the proposed scheme is able to reduce the over-segmentation while maintaining sharp region boundaries for semantically important objects.

3D Face Recognition Using Affine Integral Invariants

Abstract: A new 3D face representation and recognition approach is presented in this paper. Two sets of facial curves are extracted from a face range image, and a novel facial feature representation, the affine integral invariant, is introduced to mitigate the effect of pose on the facial curves. A human face is shown to be representable by a small subset of those affine integral invariant curves. A recognition procedure based on the discriminant analysis and Jensen-Shannon divergence analysis is proposed. Substantiating examples are provided with an achieved classification accuracy of 92.57%

Region-Based Statistical Background Modeling for Foreground Object Segmentation

Abstract: This paper proposes a novel region-based scheme for dynamically modeling time-evolving statistics of video background, leading to an effective segmentation of foreground moving objects for a video surveillance system. In (L. Li et al., 2004) statistical-based video surveillance systems employ a Bayes decision rule for classifying foreground and background changes in individual pixels. Although principal feature representations significantly reduce the size of tables of statistics, pixel-wise maintenance remains a challenge due to the computations and memory requirement. The proposed region-based scheme, which is an extension of the above method, replaces pixel-based statistics by region-based statistics through introducing dynamic background region (or pixel) merging and splitting. Simulations have been performed to several outdoor and indoor image sequences, and results have shown a significant reduction of memory requirements for tables of statistics while maintaining relatively good quality in foreground segmented video objects.

Generalization and analysis of the conventional beamformer for localization of spatially distributed sources

Abstract: In this paper, we generalize the point source-based conventional beamformer (CBF) to localization of multiple distributed sources that appear in sensor array processing. A distributed source is commonly parameterized by its mean angle and spatial spread. The generalized CBF uses the principal eigenvector of the parameterized signal covariance matrix as its optimal weight vector, which is also shown to be a matched filter. The desired parameter estimates are taken as the peaks of the generalized 2-dimensional beamforming spectrum. Further, the performance of the algorithm is compared numerically to a generalized Capon estimator [1]. Finally, an asymptotic performance analysis of the proposed algorithm is provided and numerically verified.

Hammerstein System Linearization with Application to Nonlinear Acoustic Echo Cancellation

Abstract: This paper addresses compensation for nonlinearity in Hammerstein nonlinear systems. We propose a predistortion method that is based on a coherence function criterion. The proposed method carries out linearization without knowing the linear block in the Hammerstein system. This is particularly desirable for nonlinear acoustic echo cancellation applications where dealing with the linear block can be computational cumbersome due to the long room acoustic impulse response. Effectiveness of the algorithm is demonstrated through computer simulations.

Optimal Subspace Techniques for DOA Estimation in the Presence of Noise and Model Errors

Abstract: Signal parameter estimation and specifically direction of arrival (DOA) estimation for sensor array data is encountered in a number of applications ranging from electronic surveillance to wireless communications. Subspace based methods have shown to provide computationally as well as statistically efficient algorithms for DOA estimation. Estimator performance is ultimately limited by model disturbances such as measurement noise and model errors. Herein, we review a recently proposed framework that allows the derivation of optimal subspace methods taking both finite sample effects (noise) and model perturbations into account. We show how this general estimator reduces to well known techniques for cases when one disturbance dominates completely over the other.