Support Estimation of a Sample Space-Time Covariance Matrix

Delaosa, Connor; Pestana, Jennifer; Goddard, N J; Somasundaram, S.; Weiss, Stephan

doi:10.1109/sspd.2019.8751663

Cited by 9 publications

(18 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For these, algorithmic efforts have been reviewed, with the main focus currently directed to extracting analytic eigenvalues and -vectors that afford lower-order polynomial approximations than the state-of-the-art in PEVD approaches with proven convergence. Important future developments also target the estimation of space-time covariance [63,64,65], since the statistics typically have to be estimated from finite data sets, and interesting new applications where the polynomial approach permits solutions that were previously unobtainable, such as in the area of impulse response modelling [66] or speech enhancement [67].…”

Section: Discussionmentioning

confidence: 99%

Mathematical Tools for Processing Broadband Multi-Sensor Signals

Weiss¹

2020

World Congress on Electrical Engineering and Computer Systems and Science

Self Cite

View full text Add to dashboard Cite

Spatial information in broadband array signals is embedded in the relative delay with which sources illuminate different sensors. Therefore, second order statistics, on which cost functions such as the mean square rest, must include such delays. Typically, a space-time covariance matrix therefore arises, which can be represented as a Laurent polynomial matrix. The optimisation of a cost function then requires extending the utility of the eigenvalue decomposition from narrowband covariance matrices to the broadband case of operating in a space-time covariance matrix. This overview paper summarises efforts in performing such factorisations, and demonstrated via the exemplar application of a broadband beamformer how thus well-known narrowband solutions can be extended to the broadband case using polynomial matrices and their factorisations.

show abstract

Section: Discussionmentioning

confidence: 99%

Mathematical Tools for Processing Broadband Multi-Sensor Signals

Weiss¹

2020

World Congress on Electrical Engineering and Computer Systems and Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…Assuming that the first few frames contain only the interferer components, the space-time covariance matrix in (2) can be estimated using [24], [25]. After computing the PEVD of (2), the orthogonal complement subspace U ⊥ (z) is generated according to (5).…”

Section: A Polynomial Subspace Projectionmentioning

confidence: 99%

A Polynomial Subspace Projection Approach for the Detection of Weak Voice Activity

Neo

Weiss

Naylor

2022

2022 Sensor Signal Processing for Defence Conference (SSPD)

Self Cite

View full text Add to dashboard Cite

A voice activity detection (VAD) algorithm identifies whether or not time frames contain speech. It is essential for many military and commercial speech processing applications, including speech enhancement, speech coding, speaker identification, and automatic speech recognition. In this work, we adopt earlier work on detecting weak transient signals and propose a polynomial subspace projection pre-processor to improve an existing VAD algorithm. The proposed multi-channel pre-processor projects the microphone signals onto a lower dimensional subspace which attempts to remove the interferer components and thus eases the detection of the speech target. Compared to applying the same VAD to the microphone signal, the proposed approach almost always improves the F1 and balanced accuracy scores even in adverse environments, e.g. -30 dB SIR, which may be typical of operations involving noisy machinery and signal jamming scenarios.

show abstract

“…We assume that a number of L sources have been stationary for a period of time, over which a space-time covariance matrixR[τ ] has been estimated, using e.g. the procedures outlined in [32], [33] for the estimation and the optimum support length of this estimate. Using an approximation of the PhEVD in (4) by algorithms of the second order sequential best rotation (SBR2) [14], [17] or sequential matrix diagonalisation (SMD) [18] families to factoriseR(z) • • R[τ ], we establish the broadband signal-plus-noise subspace spanned by the columns of U (z) and its complement, the noise-only subspace, spanned by the columns of U ⊥ (z), as defined in (8) with R = L.…”

Section: A Approachmentioning

confidence: 99%

Detection of Weak Transient Signals Using a Broadband Subspace Approach

Weiss

Delaosa

Matthews

et al. 2021

2021 Sensor Signal Processing for Defence Conference (SSPD)

Self Cite

View full text Add to dashboard Cite

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.

show abstract

Support Estimation of a Sample Space-Time Covariance Matrix

Cited by 9 publications

References 19 publications

Mathematical Tools for Processing Broadband Multi-Sensor Signals

Mathematical Tools for Processing Broadband Multi-Sensor Signals

A Polynomial Subspace Projection Approach for the Detection of Weak Voice Activity

Detection of Weak Transient Signals Using a Broadband Subspace Approach

Contact Info

Product

Resources

About