Noise Correlation Research Articles

Noise classification in three-level quantum networks by Machine Learning

Abstract We investigate a machine learning based classification of noise acting on a small quantum network with the aim of detecting spatial or multilevel correlations, and the interplay with Markovianity. We control a three-level system by inducing coherent population transfer exploiting different pulse amplitude combinations as inputs to train a feedforward neural network. We show that supervised learning can classify different types of classical dephasing noise affecting the system. Three non-Markovian (quasi-static correlated, anti-correlated and uncorrelated) and Markovian noises are classified with more than $99\%$ accuracy. On the contrary, correlations of Markovian noise cannot be discriminated with our method. Our approach is robust to statistical measurement errors and retains its effectiveness for physical measurements where only a limited number of samples is available making it very experimental-friendly. Our result paves the way for classifying spatial correlations of noise in quantum architectures.

The geometry of correlated variability leads to highly suboptimal discriminative sensory coding.

The brain represents the world through the activity of neural populations; however, whether the computational goal of sensory coding is to support discrimination of sensory stimuli or to generate an internal model of the sensory world is unclear. Correlated variability across a neural population (noise correlations) is commonly observed experimentally, and many studies demonstrate that correlated variability improves discriminative sensory coding compared to a null model with no correlations. However, such results do not address whether correlated variability is for discriminative sensory coding. If the computational goal of sensory coding is discriminative, then correlated variability should be optimized to support that goal. We assessed optimality of noise correlations for discriminative sensory coding in diverse datasets by developing two novel null models, each with a biological interpretation. Across datasets, we found correlated variability in neural populations leads to highly suboptimal discriminative sensory coding according to both null models. Furthermore, biological constraints prevent many subsets of the neural populations from achieving optimality, and subselecting based on biological criteria leaves discriminative coding performance suboptimal. Finally, we show that optimal subpopulation are exponentially small as the population size grows. Together, these results demonstrate that the geometry of correlated variability leads to highly suboptimal discriminative sensory coding.

Consistent fusion for distributed multi-rate multi-sensor linear systems with unknown correlated measurement noises

Consistent fusion for distributed multi-rate multi-sensor linear systems with unknown correlated measurement noises

A deep brain stimulation-conditioned RF coil for 3T MRI.

To develop and test an MRI coil assembly for imaging deep brain stimulation (DBS) at 3 T with a reduced level of local specific absorption rate of RF fields near the implant. A mechanical rotatable linearly polarized birdcage transmitter outfitted with a 32-channel receive array was constructed. The coil performance and image quality were systematically evaluated using bench-level measurements and imaging performance tests, including SNR maps, array element noise correlation, and acceleration capabilities. Electromagnetic simulations and phantom experiments were performed with clinically relevant DBS device configurations to evaluate the reduction of specific absorption rate and temperature near the implant compared with a circular polarized body coil setup. The linearly polarized birdcage coil features a block-shaped low electric field region to be co-aligned with the implanted DBS lead trajectory, while the close-fit receive array enables imaging with high SNR and enhanced encoding capabilities. The 3T coil assembly, consisting of a rotating linear birdcage and a 32-channel close-fit receive array, showed DBS-conditioned imaging technology with substantially reduced heat generation at the DBS implants.

Switching dynamics in Al/InAs nanowire-based gate-controlled superconducting switch

The observation of the gate-controlled supercurrent (GCS) effect in superconducting nanostructures increased the hopes for realizing a superconducting equivalent of semiconductor field-effect transistors. However, recent works attribute this effect to various leakage-based scenarios, giving rise to a debate on its origin. A proper understanding of the microscopic process underlying the GCS effect and the relevant time scales would be beneficial to evaluate the possible applications. In this work, we observed gate-induced two-level fluctuations between the superconducting state and normal state in Al/InAs nanowires (NWs). Noise correlation measurements show a strong correlation with leakage current fluctuations. The time-domain measurements show that these fluctuations have Poissonian statistics. Our detailed analysis of the leakage current measurements reveals that it is consistent with the stress-induced leakage current (SILC), in which inelastic tunneling with phonon generation is the predominant transport mechanism. Our findings shed light on the microscopic origin of the GCS effect and give deeper insight into the switching dynamics of the superconducting NW under the influence of the strong gate voltage.

MFCA-MICNN: a convolutional neural network with multiscale fast channel attention and multibranch irregular convolution for noise removal in dMRI

Diffusion magnetic resonance imaging (dMRI) currently stands as the foremost noninvasive method for quantifying brain tissue microstructure and reconstructing white matter fiber pathways. However, the inherent free diffusion motion of water molecules in dMRI results in signal decay, diminishing the signal-to-noise ratio (SNR) and adversely affecting the accuracy and precision of microstructural data. In response to this challenge, we propose a novel method known as the Multiscale Fast Attention-Multibranch Irregular Convolutional Neural Network for dMRI image denoising. In this work, we introduce Multiscale Fast Channel Attention, a novel approach for efficient multiscale feature extraction with attention weight computation across feature channels. This enhances the model's capability to capture complex features and improves overall performance. Furthermore, we propose a multi-branch irregular convolutional architecture that effectively disrupts spatial noise correlation and captures noise features, thereby further enhancing the denoising performance of the model. Lastly, we design a novel loss function, which ensures excellent performance in both edge and flat regions. Experimental results demonstrate that the proposed method outperforms other state-of-the-art deep learning denoising methods in both quantitative and qualitative aspects for dMRI image denoising with fewer parameters and faster operational speed.

The impact of functional correlations on task information coding

ABSTRACT State-dependent neural correlations can be understood from a neural coding framework. Noise correlations—trial-to-trial or moment-to-moment covariability—can be interpreted only if the underlying signal correlation—similarity of task selectivity between pairs of neural units—is known. Despite many investigations in local spiking circuits, it remains unclear how this coding framework applies to large-scale brain networks. Here, we investigated relationships between large-scale noise correlations and signal correlations in a multitask human fMRI dataset. We found that task-state noise correlation changes (e.g., functional connectivity) did not typically change in the same direction as their underlying signal correlation (e.g., tuning similarity of two regions). Crucially, noise correlations that changed in the opposite direction as their signal correlation (i.e., anti-aligned correlations) improved information coding of these brain regions. In contrast, noise correlations that changed in the same direction (aligned noise correlations) as their signal correlation did not. Interestingly, these aligned noise correlations were primarily correlation increases, suggesting that most functional correlation increases across fMRI networks actually degrade information coding. These findings illustrate that state-dependent noise correlations shape information coding of functional brain networks, with interpretation of correlation changes requiring knowledge of underlying signal correlations.

Study and modeling of the ultrasonic ambient noise induced by the fluid-structure interaction in a pipe

An elastic guided wave-based structural health monitoring (SHM) system consists in examining the integrity of a structure through the propagation of elastic waves. In this context, passive methods turn initial hindrances into advantages by making use of the ambient ultrasonic noise in a structure in operation, through the cross-correlation. Although they have been empirically applied in several configurations, no model is available to predict the link between the flow regime and the convergence of the ambient noise correlation towards the impulse response of the system. Considering the different parameters involved in the case of a pipe, a physical approach of this issue may be studied through the turbulent boundary layer, which has long been described by the Corcos model. Corcos-like models have essentially been applied so far on the audible range, below 1 kHz, where we are interested in frequencies around 100 kHz, typical for SHM applications. The semi-empirical model allows to determine a wall boundary pressure, which, applied at the pipe's inner boundaries, acts as a source term for ultrasonic elastic waves propagating in the pipe. Our goal is then to study if such a model can be used at these high frequencies to reproduce the experimental observations.

Periodic information enhancing and controlling stability of complex corporate financial systems

We explore the mechanism and control behavior of how periodic information affects the stability of complex corporate financial systems from two viewpoints: the corporate life cycle and the stability of financial leverage risk. A complex corporate financial system model with correlated noise and periodic driving is proposed. Furthermore, two stability indicators are proposed to measure the stability of the corporate life cycle and the risk of corporate financial leverage. Stochastic simulation results show that enhancing the intensity and frequency of controllable periodic information can improve system stability. The stability of complex corporate life cycles and financial leverage risks exhibits opposite behaviors in terms of noise correlation. In addition, it is observed that there are specific threshold values of system parameters, and the relationship with the intensity of periodic information shows non-monotonic characteristics, manifesting peculiar phenomena and transition behaviors. Finally, by comparing with actual financial data, the proposed model is verified to be superior to the benchmark model.

Adaptation of retinal discriminability to natural scenes.

Sensory systems discriminate stimuli to direct behavioral choices, a process governed by two distinct properties - neural sensitivity to specific stimuli, and stochastic properties that importantly include neural correlations. Two questions that have received extensive investigation and debate are whether visual systems are optimized for natural scenes, and whether noise correlations contribute to this optimization. However, the lack of sufficient computational models has made these questions inaccessible in the context of the normal function of the visual system, which is to discriminate between natural stimuli. Here we take a direct approach to analyze discriminability under natural scenes for a population of salamander retinal ganglion cells using a model of the retinal neural code that captures both sensitivity and stochasticity. Using methods of information geometry and generative machine learning, we analyzed the manifolds of natural stimuli and neural responses, finding that discriminability in the ganglion cell population adapts to enhance information transmission about natural scenes, in particular about localized motion. Contrary to previous proposals, noise correlations reduce information transmission and arise simply as a natural consequence of the shared circuitry that generates changing spatiotemporal visual sensitivity. These results address a long-standing debate as to the role of retinal correlations in the encoding of natural stimuli and reveal how the highly nonlinear receptive fields of the retina adapt dynamically to increase information transmission under natural scenes by performing the important ethological function of local motion discrimination.

Differential Encoding of Two-Tone Harmonics in the Male and Female Mouse Auditory Cortex.

Harmonics are an integral part of music, speech, and vocalizations of animals. Since the rest of the auditory environment is primarily made up of nonharmonic sounds, the auditory system needs to perceptually separate the above two kinds of sounds. In mice, harmonics, generally with two-tone components (two-tone harmonic complexes, TTHCs), form an important component of vocal communication. Communication by pups during isolation from the mother and by adult males during courtship elicits typical behaviors in female mice-dams and adult courting females, respectively. Our study shows that the processing of TTHC is specialized in mice providing neural basis for perceptual differences between tones and TTHCs and also nonharmonic sounds. Investigation of responses in the primary auditory cortex (Au1) from in vivo extracellular recordings and two-photon Ca2+ imaging of excitatory and inhibitory neurons to TTHCs exhibit enhancement, suppression, or no-effect with respect to tones. Irrespective of neuron type, harmonic enhancement is maximized, and suppression is minimized when the fundamental frequencies (F 0) match the neuron's best fundamental frequency (BF0). Sex-specific processing of TTHC is evident from differences in the distributions of neurons' best frequency (BF) and best fundamental frequency (BF0) in single units, differences in harmonic suppressed cases re-BF0, independent of neuron types, and from pairwise noise correlations among excitatory and parvalbumin inhibitory interneurons. Furthermore, TTHCs elicit a higher response compared with two-tone nonharmonics in females, but not in males. Thus, our study shows specialized neural processing of TTHCs over tones and nonharmonics, highlighting local network specialization among different neuronal types.

Estimation of seismometer clock time offsets using Kalman Filter towards accurate seismic velocity change

SUMMARY Monitoring seismic velocity changes obtained from ambient noise correlations is widely used to understand changes in rock properties in response to earthquakes, volcanic activities and environmental changes. Since continuous seismic data have been accumulated, this method can estimate long-term changes in seismic velocity, such as crustal recovery after a major earthquake and temporal variations in seismic velocity related to long-term environmental change. Changes in seismic velocity can be estimated with a high temporal resolution by measuring the phase differences of ambient noise correlations based on a seismic interferometry method. Still, these phase differences are influenced not only by seismic wave velocity changes but also by errors in clock timing in seismometers. The clock drift occurs due to out-of-synchronization with the GPS clock and the drift of the internal clock. Therefore, to accurately monitor temporal changes in crustal structure by measuring the phase differences of noise correlations, it is crucial to evaluate the contribution of errors in clock timing to the phase differences. Recently, a method using an extended Kalman filter based on a state-space model was developed for reliable detection of temporal changes in the waveforms of ambient noise correlations, with the state-space model offering the advantage of flexible modelling of time-series data. In this study, we incorporated the time-shifts caused by clock time errors of the seismometer into the state-space model of the temporal changes in ambient noise correlations. We estimated seismic velocity changes, amplitude changes of noise correlations and clock time errors from 2010 April to 2021 September at seismic stations around the Shinmoe-dake volcano in Japan, which experienced eruptions in 2011 and 2018, respectively. Several stations exhibited clear clock time offsets, and the occurrence of clock time-shifts coincided with the dates when the data logger was turned off for seismic station maintenance or replacement of the seismometer. The proposed method provides stable estimations with respect to the signal-to-noise ratio of the waveform, and this stable estimation facilitates accurate timing of seismic recordings, enabling precise analysis of seismic phase arrival times.

A brain structure learning-guided multi-view graph representation learning for brain network analysis.

Resting-state brain networks represent the interconnectivity of different brain regions during rest. Utilizing brain network analysis methods to model these networks can enhance our understanding of how different brain regions collaborate and communicate without explicit external stimuli. However, analyzing resting-state brain networks faces challenges due to high heterogeneity and noise correlation between subjects. This study proposes a brain structure learning-guided multi-view graph representation learning method to address the limitations of current brain network analysis and improve the diagnostic accuracy (ACC) of mental disorders. We first used multiple thresholds to generate different sparse levels of brain networks. Subsequently, we introduced graph pooling to optimize the brain network representation by reducing noise edges and data inconsistency, thereby providing more reliable input for subsequent graph convolutional networks (GCNs). Following this, we designed a multi-view GCN to comprehensively capture the complexity and variability of brain structure. Finally, we employed an attention-based adaptive module to adjust the contributions of different views, facilitating their fusion. Considering that the Smith atlas offers superior characterization of resting-state brain networks, we utilized the Smith atlas to construct the graph network. Experiments on two mental disorder datasets, the Autism Brain Imaging Data Exchange (ABIDE) dataset and the Mexican Cocaine Use Disorders (SUDMEX CONN) dataset, show that our model outperforms the state-of-the-art methods, achieving nearly 75% ACC and 70% area under the receiver operating characteristic curve (AUC) on both datasets. These findings demonstrate that our method of combining multi-view graph learning and brain structure learning can effectively capture crucial structural information in brain networks while facilitating the acquisition of feature information from diverse perspectives, thereby improving the performance of brain network analysis.

Robust recursive sigma point Kalman filtering for Huber-based generalized M-estimation

For nonlinear state estimation driven by non-Gaussian noise, the estimator is required to be updated iteratively. Since the iterative update approximates a linear process, it fails to capture the nonlinearity of observation models, and this further degrades filtering accuracy and consistency. Given the flaws of nonlinear iteration, this work incorporates a recursive strategy into generalized M-estimation rather than the iterative strategy. The proposed algorithm extends nonlinear recursion to nonlinear systems using the statistical linear regression method. The recursion allows for the gradual release of observation information and consequently enables the update to proceed along the nonlinear direction. Considering the correlated state and observation noise induced by recursions, a separately reweighting strategy is adopted to build a robust nonlinear system. Analogous to the nonlinear recursion, a robust nonlinear recursive update strategy is proposed, where the associated covariances and the observation noise statistics are updated recursively to ensure the consistency of observation noise statistics, thereby completing the nonlinear solution of the robust system. Compared with the iterative update strategies under non-Gaussian observation noise, the recursive update strategy can facilitate the estimator to achieve higher filtering accuracy, stronger robustness, and better consistency. Therefore, the proposed strategy is more suitable for the robust nonlinear filtering framework.

Effect of background noise characteristics on early warning indicators of thermoacoustic instability

In this work, we investigate the effects of background noise characteristics — specifically noise color (or correlation time) and intensity — arising from velocity-coupled and additive noise sources, on the early warning indicators (EWIs) of thermoacoustic instability. We employ a nonlinear reduced-order combustion dynamics model for our investigation. In the absence of noise, the nonlinear model undergoes a subcritical Hopf bifurcation, and our focus lies within the linearly stable region of the system (subthreshold region). The studied class of EWIs encompasses those derived from time series (variance), spectral analysis (coherence factor), Hurst exponent, and nonlinear methods (permutation entropy). We find that when the background noise is purely multiplicative, the trends in EWIs are primarily influenced by noise characteristics rather than the control parameter. Further, the EWIs cannot be estimated at low noise levels for large correlation times. In case of purely additive noise driven system, the coherence factor and variance are reliable EWIs across all range of investigated noise characteristics. The Hurst exponent can serve as effective EWI when the system features large noise correlation times, while permutation entropy is effective only when the system features small noise correlation time, i.e., where white noise assumption is acceptable. When the background noise includes contributions from both multiplicative and additive sources, coherence factor and variance emerges as the most reliable EWIs. These results provide insights for selecting appropriate EWIs to be employed in practical systems, considering potential variations in noise characteristics with simultaneous changes in combustor operating conditions.Novelty and significanceThis study conclusively shows that background noise (multiplicative and additive) characteristics – noise correlation time (color) and intensity – can significantly change the trends in early warning indicators (EWIs) for predicting the impending thermoacoustic oscillations. In gas turbine combustors, where thermoacoustic instability poses a critical challenge, inherent noise dynamics undergo variations with changing operating conditions and combustor designs. The development of effective EWIs to foresee the onset of thermoacoustic instability is crucial for preventing potential damage and ensuring the reliable operation of combustion systems. Previous works on stochastic dynamics of combustors simplify noise with the additive white noise assumption. Thus, our results demonstrated on a nonlinear combustion dynamics model advance the state-of-the-art with respect to the early prediction and control of thermoacoustic instability. The results provide valuable insights for selecting appropriate EWIs for engine monitoring in the absence of information on noise properties and their variation with operating conditions. This practical information is of direct relevance and interest to any gas turbine manufacturer/user.

Optimizing power-efficiency dynamics in ambient energy harvesting: Exploring trade-offs, linearity, and synergy

As the demand for low-power electronics and IoT devices grows, ambient energy harvesting appears to be a promising alternative for powering such systems in the long run. However, optimizing power and efficiency concurrently in such systems is challenging, involving balancing a number of variables. This paper investigates the optimization of power and efficiency in ambient energy harvesting systems focusing on nonlinear oscillator electromechanical harvesters subjected to multiplicative time-correlated ambient noise. Through extensive numerical simulations, we reveal distinct relationships between power and efficiency, influenced by various parameters. We observe autonomous stochastic resonance phenomena, elucidating a linear power-efficiency trend for small noise correlation time under fixed noise variance but limiting simultaneous power and efficiency optimization beyond a threshold. Under fixed noise strength, there is a trade-off between power and efficiency. Additionally, damping strength, piezoelectric parameters, and capacitor charging time impact power and efficiency linearly. These insights enhance understanding of power efficiency dynamics in ambient energy harvesting, thereby offering practical recommendations for parameter selection to maximize both power output and efficiency in the next generation of electronics.

A Survey on Maximum Ratio Combination: Applications, Evaluation and Future Directions

With the rapid development of wireless communications, the occupation of time and frequency resources becomes more crowded. The exploitation of space resources is necessary and the diversity combining techniques have substantial applications. Diversity combining achieves great diversity gains and improves the ability to combat multipath fading, among which the maximum ratio combining (MRC) performs as the optimal linear combining approach. However, MRC suffers from detrimental factors such as channel fading and no Gaussian noise in practical scenarios. In this paper, we focus on a comprehensive investigation of MRC. Starting from the MRC principle and system model, we summarize typical scenarios and analyze the channel fading statistics. For the influential factors, we further review related literature on channel correlation, cochannel interference (CCI) and impulsive noise. Major performance criteria and performance bounds are derived and compared. MRC confronts new developing challenges and the major development directions are reviewed. The paper finally discusses recent works and open problems for MRC applications and development. Emerging techniques such as artificial intelligence provide novel solutions for MRC performance improvements. The paper aims to present a summarized insight to assist readers in clarifying the analyzed methodology of MRC, so as to motivate new technology integration and extensive applications of advanced communication systems.

Enhancing source separation quality via optimal sensor placement in noisy environments

The paper aims to bridge a part of the gap between source separation and sensor placement studies by addressing a novel problem: “Predicting optimal sensor placement in noisy environments to improve source separation quality”. The structural information required for optimal sensor placement is modeled as the spatial distribution of source signal gains and the spatial correlation of noise. The sensor positions are predicted by optimizing two criteria as measures of separation quality, and a gradient-based global optimization method is developed to efficiently address this optimization problem. Numerical results exhibit superior performance when compared with classical sensor placement methodologies based on mutual information, underscoring the critical role of sensor placement in source separation with noisy sensor measurements. The proposed method is applied to actual electroencephalography (EEG) data to separate the P300 source components in a brain-computer interface (BCI) application. The results show that when the sensor positions are chosen using the proposed method, to reach a certain level of spelling accuracy, fewer sensors are required compared with standard sensor locations.

Subharmonic lock-in detection and its optimization for femtosecond noise correlation spectroscopy.

Although often viewed as detrimental, fluctuations carry valuable information about the physical system from which they emerge. Femtosecond noise correlation spectroscopy (FemNoC) has recently been established to probe the ultrafast fluctuation dynamics of thermally populated magnons by measurement of their amplitude autocorrelation. Subharmonic lock-in detection is the key technique in this method, allowing us to extract the pulse-to-pulse polarization fluctuations of two femtosecond optical pulse trains transmitted through a magnetic sample. Here, we present a thorough technical description of the subharmonic demodulation technique and the FemNoC measurement system. We mathematically model the data acquisition process and identify the essential parameters that critically influence the signal-to-noise ratio of the signals. Comparing the model calculations to real datasets allows validating the predicted parameter dependences and provides a means to optimize FemNoC experiments.

The impact of ambient noise sources in subsurface models estimated from noise correlation waveforms

SUMMARY Cross-correlations of seismic ambient noise are frequently used to image Earth structure. Usually, tomographic studies assume that noise sources are uniformly distributed and interpret noise correlations as empirical Green’s functions. However, previous research suggests that this assumption can introduce errors in the estimated models, especially when noise correlation waveforms are inverted. In this paper, we investigate changes in subsurface models inferred from noise correlation waveforms depending on whether the noise source distribution is considered to be uniform. To this end, we set up numerical experiments that mimic a tomographic study in Southern California exploiting ambient noise generated in the Pacific Ocean. Our results show that if the distribution of noise sources is deemed uniform instead of being numerically represented in the wave simulations, the misfit of the estimated models increases. In our experiments, the model misfit increase ranges between 5 and 21 per cent, depending on the heterogeneity of the noise source distribution. This indicates that assuming uniform noise sources introduces source-dependent model errors. Since the location of noise sources may change over time, these errors are also time-dependent. In order to mitigate these errors, it is necessary to account for the noise source distribution. The spatial extent to which noise sources must be considered depends on the propagation distance of the ambient noise wavefield. If only sources near the study area are considered, model errors may arise.