Abstract:An analytic parahermitian matrix admits an eigenvalue decomposition (EVD) with analytic eigenvalues and eigenvectors except in the case of multiplexed data. In this paper, we propose an iterative algorithm for the estimation of the analytic eigenvalues. Since these are generally transcendental, we find a polynomial approximation with a defined error. Our approach operates in the discrete Fourier transform (DFT) domain and for every DFT length generates a maximally smooth association through EVDs evaluated in D… Show more
“…The evaluation of the discrimination is thresholdindependent, and future work would have to focus on setting a suitable detection threshold for a hypothesis test on the absence or presence of a transient signal. Also, the computational burden of the processor rests on the order of the polynomial factorisation of the space-time covariance matrix, with shorter order decompositions emerging [20], [21].…”
Section: Discussionmentioning
confidence: 99%
“…Similar to narrowband subspace methods, a diagonalisation of this spacetime covariance is required to access a subspace decomposition akin to the narrowband case. For the broadband problem, we are looking towards polynomial EVD methods that can decouple the space-time covariance for every lag value [14] -such decompositions have been shown to exist in most case [15], [16] and a number of algorithms have been developed to solves this diagonalisation often with guaranteed convergence [14], [17]- [21].…”
We investigate the detection of broadband weak transient signals by monitoring a projection of the measurement data onto the noise-only subspace derived from the stationary sources. This projection utilises a broadband subspace decomposition of the data's space-time covariance matrix. The energy in this projected 'syndrome' vector is more discriminative towards the presence or absence of a transient signal than the original data, and can be enhanced by temporal averaging. We investigate the statistics, and indicate in simulations how discrimination can be traded off with the time to reach a decision, as well as with the sample size over which the space-time covariance is estimated.
“…The evaluation of the discrimination is thresholdindependent, and future work would have to focus on setting a suitable detection threshold for a hypothesis test on the absence or presence of a transient signal. Also, the computational burden of the processor rests on the order of the polynomial factorisation of the space-time covariance matrix, with shorter order decompositions emerging [20], [21].…”
Section: Discussionmentioning
confidence: 99%
“…Similar to narrowband subspace methods, a diagonalisation of this spacetime covariance is required to access a subspace decomposition akin to the narrowband case. For the broadband problem, we are looking towards polynomial EVD methods that can decouple the space-time covariance for every lag value [14] -such decompositions have been shown to exist in most case [15], [16] and a number of algorithms have been developed to solves this diagonalisation often with guaranteed convergence [14], [17]- [21].…”
We investigate the detection of broadband weak transient signals by monitoring a projection of the measurement data onto the noise-only subspace derived from the stationary sources. This projection utilises a broadband subspace decomposition of the data's space-time covariance matrix. The energy in this projected 'syndrome' vector is more discriminative towards the presence or absence of a transient signal than the original data, and can be enhanced by temporal averaging. We investigate the statistics, and indicate in simulations how discrimination can be traded off with the time to reach a decision, as well as with the sample size over which the space-time covariance is estimated.
“…While DFT-based methods are superior to time-domain approaches in terms of order of the extracted factors, the algorithms themselves do not scale well with the spatial dimension -in case of the eigenvalues [28], [32] -and the temporal dimension -in case of the eigenvectors [30], [33] of the input para-hermitian matrix. For the eigenvectors [30], [33], the computational bottleneck is a phase-smoothing operation to establish phase coherence between eigenvectors across frequency bins [28], [30], [31], which is non-convex in nature [30] and NP-hard to solve [33]. The DFT size, which is directly related to the time-domain support of the analytic eigenvectors, therefore determines the algorithm complexity.…”
Extracting analytic eigenvectors from parahermitian matrices relies on phase smoothing in the discrete Fourier transform (DFT) domain as its most expensive algorithmic component. Some algorithms require an a priori estimate of the eigenvector support and therefore the DFT length, while others iteratively increase the DFT. Thus in this document, we aim to complement the former and to reduce the computational load of the latter by estimating the time-domain support of eigenvectors. The proposed approach is validated via an ensemble of eigenvectors of known support, which the estimated support accurately matches.
“…In [20], a broadband subspace approach is used to detect weak transient signals. The approach uses an iterative polynomial matrix eigenvalue decomposition (PEVD) algorithm such as the family of second-order sequential best rotation (SBR2) [21,22] and sequential matrix diagonalization (SMD) approaches [23,24] in the time-domain or [25,26] in the frequency-domain. PEVD algorithms have been found useful for many broadband signal processing applications such as speech enhancement [27,28], source separation [29,30], source localizaton [31,32] and beamforming [33].…”
Voice activity detection (VAD) algorithms are essential for many speech processing applications, such as speaker diarization, automatic speech recognition, speech enhancement, and speech coding. With a good VAD algorithm, non-speech segments can be excluded to improve the performance and computation of these applications. In this paper, we propose a polynomial eigenvalue decompositionbased target-speaker VAD algorithm to detect unseen target speakers in the presence of competing talkers. The proposed approach uses frame-based processing to compute the syndrome energy, used for testing the presence or absence of a target speaker. The proposed approach is consistently among the best in F1 and balanced accuracy scores over the investigated range of signal to interference ratio (SIR) from -10 dB to 20 dB.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.