Source Separation Research Articles (Page 1)

Blind source separation (BSS), decomposing signals into mixed sources, remains underexplored in ictal source imaging. We evaluated its ability to extract ictal components localizing the seizure onset zone (SOZ) using routine low-density scalp EEG across multiple settings. We analyzed 20 seizures from ten patients with stereo-EEG-defined SOZs, all seizure-free after surgery or RF-thermocoagulation. BSS was computed over 10-, 30-, and 300-s periods preceding clinically and EEG-defined seizure onsets, using FastICA, InfoMax, and SOBI algorithms. Components were visually categorized as ictal or non-ictal. For each computation, the clearest ictal component was labeled best-of-BSS; one per seizure was further labeled unique when enhancing the ictal discharge. We estimated distances between components' localization and the SOZ and assessed sublobar concordance. Across 360 BSS computations (8164 components), sublobar SOZ localization occurred in 38.5% of ictal components (mean distance: 44.4mm), 48.2% of best-of-BSS (36.8mm), 56.7% of unique (28.9mm), versus 17.9% of non-ictal components (62.9mm). InfoMax computed on 300-s periods preceding clinical onset achieved the highest F-scores. Under these settings, best-of-BSS components localized the SOZ in 13/20 seizures (8/10 patients), with a median distance of 29.9mm. InfoMax algorithm incorporating longer time windows improves SOZ localization from routine scalp-EEG.

Automatic speech recognition in a scenario with multiple speakers in a reverberant space, such as a small courtroom, often requires multiple sensors. This leads to a problem of crosstalk that must be removed before the speech-to-text transcription is performed. This paper presents an algorithm intended for application in multi-speaker scenarios requiring speech-to-text transcription, such as court sessions or conferences. The proposed method uses Acoustic Vector Sensors to acquire audio streams. Speaker detection is performed using statistical analysis of the direction of arrival. This information is then used to perform source separation. Next, speakers’ activity in each channel is analyzed, and signal fragments containing direct speech and crosstalk are identified. Crosstalk is then suppressed using a dynamic gain processor, and the resulting audio streams may be passed to a speech recognition system. The algorithm was evaluated using a custom set of speech recordings. An increase in SI-SDR (Scale-Invariant Signal-to-Distortion Ratio) over the unprocessed signal was achieved: 7.54 dB and 19.53 dB for the algorithm with and without the source separation stage, respectively.

Application of the RANAS model of behavior change for source separation of domestic solid wastes

MR methods have been proposed to estimate blood oxygen saturation in cerebral tissues by exploiting the magnetic susceptibility difference between oxy and deoxyhemoglobin. These models neglect the contribution of extravascular susceptibility sources leading to inaccurate measurements. Recent approaches combined quantitative BOLD and QSM methods, but these models neglected the impact of the geometrical arrangements of extravascular sources on signal decay. Consequently, we propose a new model to unravel the magnitude and phase of MRI gradient echo sequence signal contributions of intravascular and extravascular compartments to produce blood-oxygen saturation and non-vascular susceptibility maps. This new model and previous MRI approaches are implemented in rat data acquired in healthy and stroke settings. The results are compared across anatomical regions of interest characterized by different extravascular magnetic susceptibilities. Results obtained with the new model improved estimations in myelin-rich white matter regions and delineated oxygen saturation decrease in stroke lesions after 60 min of middle cerebral artery occlusion. Characterization of extravascular susceptibility sources improved the overall results, highlighting its critical role in the translation toward pathological applications.

Targeted muscle reinnervation surgery reroutes residual nerve signals into spare muscles, enabling the recovery of neural information through electromyography (EMG). However, EMG signals are often overlapping, making the interpretation of limb functions complicated. Regenerative peripheral nerve interfaces surgically partition the nerve into individual fascicles that reinnervate specific muscle grafts, isolating distinct neural sources for precise control and interpretation of EMG signals. Here we combine targeted muscle reinnervation surgery of polyvalent nerves with a high-density microelectrode array implanted at a single site within a reinnervated muscle, and via mathematical source separation methods, we separate all neural signals that are redirected into a single muscle. In participants with upper-limb amputation, the deconvolution of EMG signals from four reinnervated muscles into motor unit spike trains revealed distinct clusters of motor neurons associated with diverse functional tasks. Our method enabled the extraction of multiple neural commands within a single reinnervated muscle, eliminating the need for surgical nerve division. This approach holds promises for enhancing control over prosthetic limbs and for understanding how the central nervous system encodes movement after reinnervation.

The difficulty in precisely extracting single-fault signatures from hydraulic pump composite faults, which stems from structural complexity and coupled multi-source vibrations, is tackled herein via a new diagnostic technique based on underdetermined blind source separation (UBSS). Utilizing sparse component analysis (SCA), the proposed method achieves blind source separation without relying on prior knowledge or multiple sensors. However, conventional SCA-based approaches are limited by their reliance on a predefined number of sources and their high sensitivity to noise. To overcome these limitations, an adaptive source number estimation strategy is proposed by integrating information–theoretic criteria into density peak clustering (DPC), enabling automatic source number determination with negligible additional computation. To facilitate this process, the short-time Fourier transform (STFT) is first employed to convert the vibration signals into the frequency domain. The resulting time–frequency points are then clustered using the integrated DPC–Bayesian Information Criterion (BIC) scheme, which jointly estimates both the number of sources and the mixing matrix. Finally, the original source signals are reconstructed through the minimum L1-norm optimization method. Simulation and experimental studies, including hydraulic pump composite fault experiments, verify that the proposed method can accurately separate mixed vibration signals and identify distinct fault components even under low signal-to-noise ratio (SNR) conditions. The results demonstrate the method’s superior separation accuracy, noise robustness, and adaptability compared with existing algorithms.

Several industrial applications rely on induction motors to carry out processes essential for product manufacturing. Speed control of an induction motor commonly requires a pulse width modulated inverter capable of driving a system with long cables, suppression of common mode voltage, reduction in common mode current, and suppression of electromagnetic interference. This paper proposes a three-phase motor drive aimed at maintaining a constant common-mode voltage. The proposed system consists of two three-phase conventional full bridge inverters connected in parallel and having as an input two separate direct current sources. The proposed system is controlled by using the space vector pulse width modulation technique. By properly designing the switching signal sequences for both converters, the common-mode voltage can be maintained constant, thereby reducing the associated common-mode current to an RMS value of 92.3 mA and enhancing the overall reliability of the system. The proposed system is validated through numerical simulations and by the implementation of an experimental prototype.

Music source separation, as a fundamental task in intelligent audio processing, plays a critical role in enhancing the performance of music generation, editing, and understanding systems. However, existing separation models often suffer from structural limitations such as reliance on a single modeling path, entangled time-frequency representations, and difficulty in adapting to heterogeneous sound sources. Furthermore, they struggle to maintain an effective balance between long-range dependency modeling and inference efficiency. To address these challenges, this paper proposes a novel dual-path state space modeling architecture, MSNet. By introducing decoupled modeling mechanisms for temporal and frequency pathways, and incorporating Mamba-based state space units for multidimensional structural parsing of audio signals, MSNet enhances selective control and structural representation in time-frequency modeling. Experimental results demonstrate that MSNet achieves state-of-the-art performance on the MUSDB18 dataset across multiple evaluation metrics. In particular, it shows superior robustness and stability when dealing with dynamically complex sources such as vocals and drums. Additionally, the model achieves a real-time factor (RTF) below 0.1 while maintaining superior separation quality, making it suitable for deployment in practical applications. This study not only demonstrates the feasibility of state space models for complex audio modeling but also introduces a new architectural paradigm for music source separation that balances accuracy and efficiency. The implementation is publicly available at: https://github.com/NMLAB8/Mamba-S-Net.

To develop a new sequence, MIMOSA, for highly efficient T1, T2, T2*, proton density (PD), and source separation quantitative susceptibility mapping (QSM). MIMOSA was developed based on 3D quantification using an interleaved Look-Locker acquisition sequence with a T2 preparation pulse (3D-QALAS) by combining 3D turbo Fast Low Angle Shot (FLASH) and multi-echo gradient echo acquisition modules with a spiral-like Cartesian trajectory to facilitate highly efficient acquisition. Simulations were performed to optimize the sequence. A multi-contrast/-slice zero-shot self-supervised learning algorithm was employed for reconstruction. The accuracy of quantitative mapping was assessed by comparing MIMOSA with 3D-QALAS and reference techniques in both ISMRM/NIST phantom and in vivo experiments. MIMOSA's acceleration capability was assessed at R = 3.3, 6.5, and 11.8 in in vivo experiments, with repeatability assessed through scan-rescan studies. Beyond the 3 T experiments, mesoscale quantitative mapping was performed at 750 μm isotropic resolution at 7 T. Simulations demonstrated that MIMOSA achieved improved parameter estimation accuracy compared to 3D-QALAS. Phantom experiments indicated that MIMOSA exhibited better agreement with the reference techniques than 3D-QALAS. In vivo experiments demonstrated that an acceleration factor of up to R = 11.8-fold can be achieved while preserving parameter estimation accuracy, with intra-class correlation coefficients of 0.998 (T1), 0.973 (T2), 0.947 (T2*), 0.992 (QSM), 0.987 (paramagnetic susceptibility), and 0.977 (diamagnetic susceptibility) in scan-rescan studies. Whole-brain T1, T2, T2*, PD, and source separation QSM were obtained with 1 mm isotropic resolution in 3 min at 3 T and 750 μm isotropic resolution in 13 min at 7 T. MIMOSA demonstrated potential for highly efficient and repeatable multi-parametric mapping.

Abstract Music identification is crucial for distributing royalties in the music industry. This problem is solved using Audio fingerprinting (AFP) algorithms. However, these methods often struggle in real-world scenarios such as TV broadcasting, when music is in the background, masked by other sounds such as speech. While prior research has focused on improving AFP robustness to pitch and tempo variations, less attention has been given to enhancing robustness for background music identification. In this work, we assess whether source separation systems improve background music identification by recovering the music signal in these recordings. We present the first extensive study comprising 13 source separation algorithms and five AFP models. We evaluate them on a public dataset of TV recordings, assessing both music identification performance and computational cost. Our results show that source separation substantially improves peak-based AFP identifications, particularly when music is in the background. Additionally, this finding extends to foreground music, making the approach versatile for various music identification tasks, such as query-by-example. Deep learning-based model NeuralFP* (tailored for background music identification) shows no substantial benefit from adding a separation model as preprocessing. This reproducible study provides a comprehensive evaluation framework, offering valuable insights into using source separation methods to improve music identification in real-world contexts.

Mainlobe interference can severely degrade the performance of distributed phased-array radar systems in the presence of strong jamming or low-reflectivity targets. This paper introduces a signal–data dual-domain cooperative antijamming and localization (SDCAL) framework that integrates adaptive complete ensemble empirical mode decomposition with improved blind source separation and wavelet optimization (CEEMDAN-WOBSS) for signal-level denoising and separation. Following source separation, CFAR-based pulse compression is applied for precise range estimation, and multi-node data fusion is then used to achieve three-dimensional target localization. Under low signal-to-noise ratio (SNR) conditions, the adaptive CEEMDAN–WOBSS approach reconstructs the signal covariance matrix to preserve subspace rank, thereby accelerating convergence of the separation matrix. The subsequent pulse compression and CFAR detection steps provide reliable inter-node distance measurements for accurate fusion. The simulation results demonstrate that, compared to conventional blind-source-separation methods, the proposed framework markedly enhances interference suppression, detection probability, and localization accuracy—validating its effectiveness for robust collaborative sensing in challenging jamming scenarios.

Assessment of linking perceived benefits to source separation behavior and waste management economics by Monte Carlo simulation

With the advent of artificial intelligence, various computer-aided diagnostic systems are being developed to assist medical professionals. Deep learning techniques powered by convolutional neural networks seem promising for obtaining new insights into the onco-histopathology domain. Breast cancer is confirmed by histopathological analysis of Hematoxylin and Eosin (H&E)-stained breast tissue images. Finding the molecular subtype of breast cancer using Immunohistochemistry (IHC)-stained breast tissue is essential to decide on the correct treatment plan for a breast cancer patient. IHC staining is an expensive process that is very time-consuming and involves intra- and inter-observer subjectivity. This work attempts to find the Human Epidermal growth factor Receptor Two (HER2) molecular subtype from H&E-stained tissue images instead of using IHC-stained tissues. H&E-stained tissue image data from two separate sources are used to predict HER2 status. This study aimed to improve the accuracy of HER2 overexpression classification by modifying the architecture of the ResNet50 model by cascading it with a squeeze and excitation block and a depth-wise separable convolutional layer. The dataset comprises a combination of tissue image patches from a publicly available Warwick dataset and a real-world dataset collected from a hospital in Pune, India. The dataset is preprocessed and split into 60% train, 20% validation, and 20% test subsets. The proposed architecture with a modified ResNet50 network achieves the best patch-level HER2 classification accuracy of 98.04%, with class-specific test accuracy results for HER2 negative, HER2 equivocal, and HER2 positive scores being 97.73%, 99.70%, and 98.93%, respectively.

Correction: TFSWA-ResUNet: music source separation with time–frequency sequence and shifted window attention-based ResUNet

This paper proposes a radar mainlobe anti-jamming method based on Space-Time Coding (STC) and Blind Source Separation (BSS). Addressing the performance degradation issue of traditional BSS methods under low Signal-to-Noise Ratio (SNR) and insufficient spatial resolution, this study first establishes the airborne SAR imaging geometric model and the jamming signal mixing model. Subsequently, STC technology is introduced to construct more equivalent phase centers and increase the system’s spatial Degrees of Freedom (DOF). Leveraging the increased DOFs, a JADE-based blind source separation algorithm is then employed to separate the mixed jamming signals. The separation of these signals significantly enhances the anti-jamming capability of the radar system. Simulation results demonstrate that the proposed method effectively improves BSS performance. As compared to traditional BSS schemes, this method provides an additional jamming suppression gain of approximately 10 dB in point target scenarios and about 3 dB in distributed target scenarios, significantly enhancing the radar system’s mainlobe anti-jamming capability in complex jamming environments. This method provides a new insight into radar mainlobe anti-jamming by combining the STC scheme and BSS technology.

Time domain source separation of microphone array signals with non-stationary, impulsive sources using robust principal component analysis (RPCA) is presented. Time domain source separation with RPCA is applied to microphone array signals observing the aeroacoustic emission of a vortex ring interacting with the edge of a semi-infinite half-plane (V/E interaction). An impulsive spherical pressure wave is produced as a by-product of the generation of vortex rings and is observed by all microphones. With non-stationary, impulsive signal features, frequency domain source separation techniques may not sufficiently separate the sources, requiring a time domain approach. Source separation is achieved with RPCA, enabling accurate estimation of V/E source parameters, with theoretical predictions in good agreement. RPCA, in this application, shows improved performance when compared to other time domain source separation methods involving principal component analysis (PCA). RPCA provides a data-driven approach for impulsive source separation, requiring less user intervention than PCA. Furthermore, RPCA source separation enables improved V/E source waveform time series when compared to prior efforts, which utilized signal windows that excluded the impulsive pressure wave signal feature.

Low-carbon competitiveness of cities in solid waste disposal systems: Spatial and temporal variations in greenhouse gas emissions in the Yangtze River Delta.

The article analyzes the play, which went almost unnoticed in critical thought contemporary to Gogol. It contains eight dramatic scenes united under the name of “Lackeyskaya.” A whole stream of misunderstandings, from important to curious, that arose during the perceptionandinterpretation of “Lackey”insubsequentliterature is considered. For the first time, a holisticstudy of the author’s idea of the play was proposed. The dramatic scenes are examined as the focus of the socio-political views of the writer, who studies the interrelated psychology of servility and lordship. The author’s views on national life are summarized along with the influence of Christian hagiography on Gogol. An overview of the literary works that influenced the creation of “Lackeyskaya” is provided. The article highlights the constant nature of the socia lproblems raised in the play for all of Gogol’s work; the educational, “parable-like” (in Gogol’s understanding of the word) idea of “Lackeyskaya” and its autobiographical subtext are noted. Special attention is paid to the religious scope of the playand the polemical reflection in it of the ideas of the Scottish economist A. Smith on the industrial division of labor. An article by an unknown author on the position of serfs in Little Russia, preserved in the writer’s papers, was used as a separate source that served as material for Gogol’s creation of images of servants. In an “unexpected” way, the “inconspicuous” “Lackeyskaya” finds itself in the midst of Gogol’s years-long deliberations. The main result of Gogol’s artistic analysis is summarized in the conclusion that a person finds his dignity in the Church and in the service of God, whereas a society without God turns a person into a “lackey.”

Abstract Measurements and analyses of atmospheric electric fields in the frequency range from DC to a few hertz are employed in Earth and atmospheric sciences to evaluate cloud electrification, thunderstorm evolution, near-surface sources of charge production and separation, and the global electric circuit (GEC). Most operational and research uses of these data are limited by the difficulty in obtaining measurements that are minimally impacted by nearby low-mobility electrically charged particles (space charge) and near-surface radioactive sources. One such operational use is by U.S.-based space launch facilities to avoid lightning strikes to a launch vehicle during ascent. Ground-based electric field measurements are a key part of these rules, with assessments being limited by nearby “noise sources.” Additionally, GEC research is seriously hampered by the difficulty in obtaining low-noise measurements of the fair-weather electric field for individual days. In this work, we discuss our efforts to improve atmospheric electric field observations by placing an electric field mill (EFM) atop a 90-m tower, compensating for the field enhancement produced by this mounting geometry, and comparing its calibrated field measurements to those measured at three nearby ground-mounted EFMs at Kennedy Space Center. The tower installation exhibited only benefits, relative to nearby surface-mounted sites, in the form of reduced sensitivity to wind and near-surface space charge sources, reduced influence of the “sunrise effect,” reduced sensitivity to charged precipitation during warm rain, and insensitivity to nearby vegetation growth. Measurements of electric fields produced by nearby thunderstorms were not compromised by this tower-mounted configuration. Significance Statement Slowly varying electric fields in the atmosphere can be used to evaluate thunderstorm electrical charges and the daily electric field variations that they produce at remote locations, as part of the global electric circuit. Most operational and research uses of these data are limited by the difficulty in obtaining measurements that are minimally impacted by nearby noise and sources of error. One important operational use is by space launch facilities that are required to follow strict launch rules to avoid intercepting or triggering lightning by a launch vehicle during ascent. In this work, we discuss the significant benefits of making these measurements atop a 90-m tower and comparing them to fields measured on the ground at Kennedy Space Center.

Common method variance (CMV) substantially impacts how scholars conduct and review research. Several procedural and statistical remedies have been proposed to address the potential biasing effects that can result from CMV in data procured from a single source on a single occasion. Among them, temporal separation and distinct source designs have been the most popular. Psychological separation (PS) has also been proposed as a way to address CMV, by diverting respondents' attention from previously accessed memories, disrupting response consistency patterns, and improving effortful responding. The present research attempted to create efficacious PS through a cognitive interference task administered midway through a survey, thereby attenuating correlations that could be affected by CMV to varying degrees. In an initial study and a constructive replication, our results show that a PS intervention of at least 7.5-min attenuated several relationships to levels significantly lower than those in a single source on a single occasion design, but to an extent consistent with the attenuation achieved by temporal separation or distinct source designs. These findings suggest that under appropriate circumstances, PS is an effective strategy to address certain forms of CMV. We conclude by providing a decision guide for responsibly choosing a research design in light of various theoretical, methodological, and logistical considerations, as well as offering several additional PS task examples that can be deployed in future studies. (PsycInfo Database Record (c) 2025 APA, all rights reserved).