Search

A new augmented mean squared error (MSE) criterion called the Error Whitening Criterion (EWC) is proposed and analyzed. The analytical solution of the EWC is determined and stochastic gradient and other fast algorithms are developed to calculate the EWC solution in an online fashion.

We propose the use of nonnegative matrix factorization (NMF) as a model-independent methodology to analyze neural activity. We demonstrate that, using this technique, it is possible to identify local spatiotemporal patterns of neural activity in the form of sparse basis vectors. In addition, the sparseness of these bases can help infer correlations between cortical firing patterns and behavior. We demonstrate the utility of this approach using neural recordings collected in a brain-machine interface (BMI) setting. The results indicate that, using the NMF analysis, it is possible to improve the performance of BMI models through appropriate pruning of inputs.

SIPEX-G is a fast-converging, robust, gradient-based PCA algorithm that has been recently proposed by the authors. Its superior performance in synthetic and real data compared with its benchmark counterparts makes it a viable alternative in applications where subspace methods are employed. Blind multiuser detection is one such area, where subspace methods, recently developed by researchers, have proven effective. In this paper, the SIPEX-G algorithm is presented in detail, convergence proofs are derived, and the performance is demonstrated in standard subspace problems. These sub space problems include direction of arrival estimation for incoming signals impinging on a linear array of sensors, nonstationary random process subspace tracking, and adaptive blind multiuser detection.

SIPEX-G is a fast-converging, robust, gradient-based PCA algorithm that has been recently proposed by the authors. Its superior performance in synthetic and real data compared with its benchmark counterparts makes it a viable alternative in applications where subspacemethods are employed. Blindmultiuser detection is one such area, where subspacemethods, recently developed by researchers, have proven effective. In this paper, the SIPEX-G algorithm is presented in detail, convergence proofs are derived, and the performance is demonstrated in standard subspace problems. These subspace problems include direction of arrival estimation for incoming signals impinging on a linear array of sensors, nonstationary random process subspace tracking, and adaptive blind multiuser detection. [ABSTRACT FROM AUTHOR]