601 results found
Stanković L, Brajović M, Daković M, et al., 2020, On the decomposition of multichannel nonstationary multicomponent signals, Signal Processing, Vol: 167, ISSN: 0165-1684
© 2019 Elsevier B.V. With their ability to cater for simultaneously for multifaceted information, multichannel (multivariate) signals have been used to solve problems that are normally not solvable with signals obtained from a single source. One such problem is the decomposition of signals which comprise several components for which the domains of support significantly overlap in both the time, frequency and the joint time-frequency domain. Earlier, we proposed a solution to this problem based on the Wigner distribution of multichannel signals, which requires the attenuation of the cross-terms. In this paper, an advanced solution is proposed, based on eigenvalue analysis of the multichannel signal autocorrelation matrix, followed by the minimization of their time-frequency concentration measure. The analysis offers less restrictive conditions for the signal decomposition, compared to the case of the Wigner distribution. The algorithm for the separation of components is based on concentration measures of the eigenvector time-frequency representation, which represent linear combinations of the overlapping signal components. With an increased number of sensors/channels, the robustness of the decomposition process to additive noise is also demonstrated. The theory is supported by numerical examples, whereby the required channel dissimilarity is also statistically investigated.
Solo V, Greco MS, Mandic D, et al., 2019, Innovation Starts With Education ICASSP 2019 Education Panel, IEEE SIGNAL PROCESSING MAGAZINE, Vol: 36, Pages: 135-+, ISSN: 1053-5888
Mandic DP, Djuric PM, Cichocki A, et al., 2019, Quo Vadis ICASSP: Echoes of 2019 ICASSP in Brighton, United Kingdom Signal Processing Meets the Needs of Modern Humankind, IEEE SIGNAL PROCESSING MAGAZINE, Vol: 36, Pages: 127-134, ISSN: 1053-5888
Talebi SP, Werner S, Mandic DP, 2019, Complex-Valued Nonlinear Adaptive Filters With Applications in alpha-Stable Environments, IEEE SIGNAL PROCESSING LETTERS, Vol: 26, Pages: 1315-1319, ISSN: 1070-9908
Xiang M, Xia Y, Mandic DP, 2019, Complementary Cost Functions for Complex and Quaternion Widely Linear Estimation, IEEE SIGNAL PROCESSING LETTERS, Vol: 26, Pages: 1344-1348, ISSN: 1070-9908
Phan A-H, Cichocki A, Oseledets I, et al., 2019, Tensor Networks for Latent Variable Analysis: Higher Order Canonical Polyadic Decomposition., IEEE Trans Neural Netw Learn Syst
The canonical polyadic decomposition (CPD) is a convenient and intuitive tool for tensor factorization; however, for higher order tensors, it often exhibits high computational cost and permutation of tensor entries, and these undesirable effects grow exponentially with the tensor order. Prior compression of tensor in-hand can reduce the computational cost of CPD, but this is only applicable when the rank $R$ of the decomposition does not exceed the tensor dimensions. To resolve these issues, we present a novel method for CPD of higher order tensors, which rests upon a simple tensor network of representative inter-connected core tensors of orders not higher than 3. For rigor, we develop an exact conversion scheme from the core tensors to the factor matrices in CPD and an iterative algorithm of low complexity to estimate these factor matrices for the inexact case. Comprehensive simulations over a variety of scenarios support the proposed approach.
Oliveira V, von Rosenberg W, Montaldo P, et al., 2019, Early postnatal heart rate variability in healthy newborn infants, Frontiers in Physiology, Vol: 10, Pages: 1-12, ISSN: 1664-042X
Background: Despite the increasing interest in fetal and neonatal heart rate variability (HRV) analysis and its potential use as a tool for early disease stratification, no studies have previously described the normal trends of HRV in healthy babies during the first hours of postnatal life.Methods: We prospectively recruited 150 healthy babies from the postnatal ward and continuously recorded their electrocardiogram during the first 24 h after birth. Babies were included if born in good condition and stayed with their mother. Babies requiring any medication or treatment were excluded. Five-minute segments of the electrocardiogram (non-overlapping time-windows) with more than 90% consecutive good quality beats were included in the calculation of hourly medians and interquartile ranges to describe HRV trends over the first 24 h. We used multilevel mixed effects regression with auto-regressive covariance structure for all repeated measures analysis and t-tests to compare group differences. Non-normally distributed variables were log-transformed.Results: Nine out of 16 HRV metrics (including heart rate) changed significantly over the 24 h [Heart rate p < 0.01; Standard deviation of the NN intervals p = 0.01; Standard deviation of the Poincaré plot lengthwise p < 0.01; Cardiac sympathetic index (CSI) p < 0.01; Normalized high frequency power p = 0.03; Normalized low frequency power p < 0.01; Total power p < 0.01; HRV index p = 0.01; Parseval index p = 0.03], adjusted for relevant clinical variables. We observed an increase in several HRV metrics during the first 6 h followed by a gradual normalization by approximately 12 h of age. Between 6 and 12 h of age, only heart rate and the normalized low frequency power changed significantly, while between 12 and 18 h no metric, other than heart rate, changed significantly. Analysis with multilevel mixed effects regression analysis (multivariable) revealed that gestational age, reduced fetal movements, cardi
Zhang X, Xia Y, Li C, et al., 2019, Analysis of the Unconstrained Frequency-Domain Block LMS for Second-Order Noncircular Inputs, IEEE TRANSACTIONS ON SIGNAL PROCESSING, Vol: 67, Pages: 3970-3984, ISSN: 1053-587X
Xia Y, Tao S, Li Z, et al., 2019, Full Mean Square Performance Bounds on Quaternion Estimators for Improper Data, IEEE TRANSACTIONS ON SIGNAL PROCESSING, Vol: 67, Pages: 4093-4106, ISSN: 1053-587X
Nitta T, Kobayashi M, Mandic DP, 2019, Hypercomplex Widely Linear Estimation Through the Lens of Underpinning Geometry, IEEE TRANSACTIONS ON SIGNAL PROCESSING, Vol: 67, Pages: 3985-3994, ISSN: 1053-587X
Stott AE, Dees BS, Kisil I, et al., 2019, A class of multidimensional NIPALS algorithms for quaternion and tensor partial least squares regression, SIGNAL PROCESSING, Vol: 160, Pages: 316-327, ISSN: 0165-1684
Powezka K, Adjei T, von Rosenberg W, et al., 2019, A pilot study of preoperative heart rate variability predicting pain during local anesthetic varicose vein surgery, JOURNAL OF VASCULAR SURGERY-VENOUS AND LYMPHATIC DISORDERS, Vol: 7, Pages: 382-386, ISSN: 2213-333X
Li Z, Xia Y, Pei W, et al., 2019, A cost-effective nonlinear self-interference canceller in full-duplex direct-conversion transceivers, SIGNAL PROCESSING, Vol: 158, Pages: 4-14, ISSN: 0165-1684
Kanna S, Moniri A, Xia Y, et al., 2019, A Data Analytics Perspective of Power Grid Analysis-Part 2: Teaching Old Power Systems New Tricks, IEEE SIGNAL PROCESSING MAGAZINE, Vol: 36, Pages: 110-117, ISSN: 1053-5888
Adjei T, von Rosenberg W, Nakamura T, et al., 2019, The ClassA framework: HRV based assessment of SNS and PNS dynamics without LF-HF controversies, Frontiers in Physiology, Vol: 10, ISSN: 1664-042X
The powers of the low frequency (LF) and high frequency (HF) components of heart rate variability (HRV) have become the de facto standard metrics in the assessment of the stress response, and the related activities of the sympathetic nervous system (SNS) and the parasympathetic nervous system (PNS). However, the widely adopted physiological interpretations of the LF and HF components in SNS /PNS balance are now questioned, which puts under serious scrutiny stress assessments which employ the LF and HF components. To avoid these controversies, we here introduce the novel Classification Angle (ClassA) framework, which yields a family of metrics which quantify cardiac dynamics in three-dimensions. This is achieved using a finite-difference plot of HRV, which displays successive rates of change of HRV, and is demonstrated to provide sufficient degrees of freedom to determine cardiac deceleration and/or acceleration. The robustness and accuracy of the novel ClassA framework is verified using HRV signals from ten males, recorded during standardized stress tests, consisting of rest, mental arithmetic, meditation, exercise and further meditation. Comparative statistical testing demonstrates that unlike the existing LF-HF metrics, the ClassA metrics are capable of distinguishing both the physical and mental stress epochs from the epochs of no stress, with statistical significance (Bonferroni corrected p-value ≤ 0.025); HF was able to distinguish physical stress from no stress, but was not able to identify mental stress. The ClassA results also indicated that at moderate levels of stress, the extent of parasympathetic withdrawal was greater than the extent of sympathetic activation. Finally, the analyses and the experimental results provide conclusive evidence that the proposed nonlinear approach to quantify cardiac activity from HRV resolves three critical obstacles to current HRV stress assessments: (i) it is not based on controversial assumptions of balance between the
Nakamura T, Alqurashi Y, Morrell M, et al., 2019, Hearables: Automatic overnight sleep monitoring with standardised in-ear EEG sensor, IEEE Transactions on Biomedical Engineering, ISSN: 0018-9294
Objective: Advances in sensor miniaturisation and computational power have served as enabling technologies for monitoring human physiological conditions in real-world scenarios. Sleep disruption may impact neural function, and can be a symptom of both physical and mental disorders. This study proposes wearable in-ear electroencephalography (ear- EEG) for overnight sleep monitoring as a 24/7 continuous and unobtrusive technology for sleep quality assessment in the community. Methods: Twenty-two healthy participants took part in overnight sleep monitoring with simultaneous ear-EEG and conventional full polysomnography (PSG) recordings. The ear- EEG data were analysed in the both structural complexity and spectral domains; the extracted features were used for automatic sleep stage prediction through supervised machine learning, whereby the PSG data were manually scored by a sleep clinician. Results: The agreement between automatic sleep stage prediction based on ear-EEG from a single in-ear sensor and the hypnogram based on the full PSG was 74.1% in the accuracy over five sleep stage classification; this is supported by a Substantial Agreement in the kappa metric (0.61). Conclusion: The in-ear sensor is both feasible for monitoring overnight sleep outside the sleep laboratory and mitigates technical difficulties associated with scalp-EEG. It therefore represents a 24/7 continuously wearable alternative to conventional cumbersome and expensive sleep monitoring. Significance: The ‘standardised’ one-size-fits-all viscoelastic in-ear sensor is a next generation solution to monitor sleep - this technology promises to be a viable method for readily wearable sleep monitoring in the community, a key to affordable healthcare and future eHealth.
Phan A-H, Yamagishi M, Mandic D, et al., Quadratic programming over ellipsoids with applications to constrained linear regression and tensor decomposition, Neural Computing and Applications, ISSN: 0941-0643
Nakamura T, Davies H, Mandic D, Scalable automatic sleep staging in the era of Big Data, IEEE EMBC 2019, Publisher: IEEE
Numerous automatic sleep staging approacheshave been proposed to provide an eHealth alternative to thecurrent gold-standard – hypnogram scoring by human experts.However, a majority of such studies exploit data of limited scale,which compromises both the validation and the reproducibilityand transferability of such automatic sleep staging systemsin the real clinical settings. In addition, the computationalissues and physical meaningfulness of the analysis are typicallyneglected, yet affordable computation is a key criterion inBig Data analytics. To this end, we establish a comprehensiveanalysis framework to rigorously evaluate the feasibility ofautomatic sleep staging from multiple perspectives, includingrobustness with respect to the number of training subjects,model complexity, and different classifiers. This is achievedfor a large collection of publicly accessible polysomnography(PSG) data, recorded over 515 subjects. The trade-off betweenaffordable computation and satisfactory accuracy is shown tobe fulfilled by an extreme learning machine (ELM) classifier,which in conjunction with the physically meaningful hiddenMarkov model (HMM) of the transition between the differentsleep stages (smoothing model) is shown to achieve both fastcomputation and highest average Cohen’s kappa value ofκ=0.73(Substantial Agreement). Finally, it is shown thatfor accurate and robust automatic sleep staging, a combinationof structural complexity (multi-scale entropy) and frequency-domain (spectral edge frequency) features is both computation-ally affordable and physically meaningful.
Mandic D, Took CC, 2019, Reply to "Comments on 'The Quaternion LMS Algorithm for Adaptive Filtering of Hypercomplex Processes"', IEEE TRANSACTIONS ON SIGNAL PROCESSING, Vol: 67, Pages: 1959-1959, ISSN: 1053-587X
Hansen ST, Hemakom A, Safeldt MG, et al., 2019, Unmixing oscillatory brain activity by EEG source localization and empirical mode decomposition, Computational Intelligence and Neuroscience, Vol: 2019, ISSN: 1687-5265
Neuronal activity is composed of synchronous and asynchronous oscillatory activity at different frequencies. The neuronal oscillations occur at time scales well matched to the temporal resolution of electroencephalography (EEG); however, to derive meaning from the electrical brain activity as measured from the scalp, it is useful to decompose the EEG signal in space and time. In this study, we elaborate on the investigations into source-based signal decomposition of EEG. Using source localization, the electrical brain signal is spatially unmixed and the neuronal dynamics from a region of interest are analyzed using empirical mode decomposition (EMD), a technique aimed at detecting periodic signals. We demonstrate, first in simulations, that the EMD is more accurate when applied to the spatially unmixed signal compared to the scalp-level signal. Furthermore, on EEG data recorded simultaneously with transcranial magnetic stimulation (TMS) over the hand area of the primary motor cortex, we observe a link between the peak to peak amplitude of the motor-evoked potential (MEP) and the phase of the decomposed localized electrical activity before TMS onset. The results thus encourage combination of source localization and EMD in the pursuit of further insight into the mechanisms of the brain with respect to the phase and frequency of the electrical oscillations and their cortical origin.
Mandic DP, Kanna S, Xia Y, et al., 2019, A Data Analytics Perspective of Power Grid Analysis-Part 1: The Clarke and Related Transforms, IEEE SIGNAL PROCESSING MAGAZINE, Vol: 36, Pages: 110-116, ISSN: 1053-5888
Moniri A, Kisil I, Constantinides AG, et al., 2019, Refreshing DSP Courses through Biopresence in the Curriculum: A Successful Paradigm, 23rd IEEE International Conference on Digital Signal Processing (DSP), Publisher: IEEE, ISSN: 1546-1874
Li Z, Deng W, Pei W, et al., 2019, Refreshing Digital Communications Curriculum with RFID Technology: A Participatory Approach, 23rd IEEE International Conference on Digital Signal Processing (DSP), Publisher: IEEE, ISSN: 1546-1874
Cheng H, Xia Y, Huang Y, et al., 2019, Joint Channel Estimation and Tx/Rx I/Q Imbalance Compensation for GFDM Systems, IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, Vol: 18, Pages: 1304-1317, ISSN: 1536-1276
Yu Z, Li S, Mandic D, 2019, Widely Linear Complex-Valued Autoencoder: Dealing with Noncircularity in Generative-Discriminative Models, Pages: 339-350, ISSN: 0302-9743
© 2019, Springer Nature Switzerland AG. We propose a new structure for the complex-valued autoencoder by introducing additional degrees of freedom into its design through a widely linear (WL) transform. The corresponding widely linear backpropagation algorithm is also developed using the (Formula Presented) calculus, to unify the gradient calculation of the cost function and the underlying WL model. More specifically, all the existing complex-valued autoencoders employ the strictly linear transform, which is optimal only when the complex-valued outputs of each network layer are independent of the conjugate of the inputs. In addition, the widely linear model which underpins our work allows us to consider all the second-order statistics of inputs. This provides more freedom in the design and enhanced optimization opportunities, as compared to the state-of-the-art. Furthermore, we show that the most widely adopted cost function, i.e., the mean squared error, is not best suited for the complex domain, as it is a real quantity with a single degree of freedom, while both the phase and the amplitude information need to be optimized. To resolve this issue, we design a new cost function, which is capable of controlling the balance between the phase and the amplitude contribution to the solution. The experimental results verify the superior performance of the proposed autoencoder together with the new cost function, especially for the imaging scenarios where the phase preserves extensive information on edges and shapes.
Xiang M, Dees BS, Mandic DP, 2019, Multiple-Model Adaptive Estimation for 3-D and 4-D Signals: A Widely Linear Quaternion Approach, IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, Vol: 30, Pages: 72-84, ISSN: 2162-237X
von Rosenberg W, Hoting MO, Mandic DP, 2019, A physiology based model of heart rate variability, Biomedical Engineering Letters, ISSN: 2093-9868
© 2019, Korean Society of Medical and Biological Engineering. Heart rate variability (HRV) is governed by the autonomic nervous system (ANS) and is routinely used to estimate the state of body and mind. At the same time, recorded HRV features can vary substantially between people. A model for HRV that (1) correctly simulates observed HRV, (2) reliably functions for multiple scenarios, and (3) can be personalised using a manageable set of parameters, would be a significant step forward toward understanding individual responses to external influences, such as physical and physiological stress. Current HRV models attempt to reproduce HRV characteristics by mimicking the statistical properties of measured HRV signals. The model presented here for the simulation of HRV follows a radically different approach, as it is based on an approximation of the physiology behind the triggering of a heart beat and the biophysics mechanisms of how the triggering process—and thereby the HRV—is governed by the ANS. The model takes into account the metabolisation rates of neurotransmitters and the change in membrane potential depending on transmitter and ion concentrations. It produces an HRV time series that not only exhibits the features observed in real data, but also explains a reduction of low frequency band-power for physically or psychologically high intensity scenarios. Furthermore, the proposed model enables the personalisation of input parameters to the physiology of different people, a unique feature not present in existing methods. All these aspects are crucial for the understanding and application of future wearable health.
Variddhisai T, Xiang M, Douglas SC, et al., 2019, Quaternion-Valued Adaptive Filtering via Nesterov's Extrapolation, 44th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Publisher: IEEE, Pages: 4868-4872, ISSN: 1520-6149
Moniri A, Constantinides AG, Mandic DP, 2019, SMART DSP FOR A SMARTER POWER GRID: TEACHING POWER SYSTEM ANALYSIS THROUGH SIGNAL PROCESSING, 44th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Publisher: IEEE, Pages: 7883-7887, ISSN: 1520-6149
Calvi GG, Lucic V, Mandic DP, 2019, SUPPORT TENSOR MACHINE FOR FINANCIAL FORECASTING, 44th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Publisher: IEEE, Pages: 8152-8156, ISSN: 1520-6149
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.