Additive Noise Research Articles

Mitigating Adversarial Attacks in Object Detection through Conditional Diffusion Models

The field of object detection has witnessed significant advancements in recent years, thanks to the remarkable progress in artificial intelligence and deep learning. These breakthroughs have significantly enhanced the accuracy and efficiency of detecting and categorizing objects in digital images. Nonetheless, contemporary object detection technologies have certain limitations, such as their inability to counter white-box attacks, insufficient denoising, suboptimal reconstruction, and gradient confusion. To overcome these hurdles, this study proposes an innovative approach that uses conditional diffusion models to perturb adversarial examples. The process begins with the application of a random chessboard mask to the adversarial example, followed by the addition of a slight noise to fill the masked area during the forward process. The adversarial image is then restored to its original form through a reverse generative process that only considers the masked pixels, not the entire image. Next, we use the complement of the initial mask as the mask for the second stage to reconstruct the image once more. This two-stage masking process allows for the complete removal of global disturbances and aids in image reconstruction. In particular, we employ a conditional diffusion model based on a class-conditional U-Net architecture, with the source image further conditioned through concatenation. Our method outperforms the recently introduced HARP method by 5% and 6.5% in mAP on the COCO2017 and PASCAL VOC datasets, respectively, under non-APT PGD attacks. Comprehensive experimental results confirm that our method can effectively restore adversarial examples, demonstrating its practical utility.

MD-BiMamba: An aero-engine inter-shaft bearing fault diagnosis method based on Mamba with modal decomposition and bidirectional features fusion strategy

Inter-shaft bearing fault diagnosis can greatly save maintenance costs and guarantee the normal operation of aero-engines. However, existing methods are limited in their ability to handle sensor signals with high dimensional and complex noise. To tackle the above problems, we propose a Mamba network structure with Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and bidirectional feature fusion for aero-engine inter-shaft bearing fault diagnosis, called MD-BiMamba. Firstly, we utilize the CEEMDAN technique to decompose the sensor signals of inter-shaft bearings and extract the intrinsic modal function by improving the noise addition method and decomposition strategy. Then, bidirectional Mamba is designed to extract the bidirectional long-range correlation of the sequence to achieve global feature extraction and fast detection of fault types with linear complexity. Moreover, an adaptive attention fusion method is proposed to achieve bidirectional feature fusion. Finally, we use the label smoothing regularization strategy to optimize the cross-entropy loss and enhance the generalization performance. Experimental results on real inter-shaft bearing datasets and noisy datasets show that the proposed method achieves 99.88 % and 95.81 % accuracy in the standard dataset and −10 dB noise environment, respectively, which is superior to other remarkable methods, in addition to achieving lower model complexity.

Inhomogeneous Hegselmann–Krause models with two types of noise

We derive two types of stochastic Hegselmann–Krause opinion formation models with leadership, study their different asymptotic behaviors and the noise effect to the leader’s control. Firstly, we propose a stochastic model with a multiplicative noise, which mimics the randomness from communication uncertainty. Using the Lyapunov functional approach, we provide a sufficient condition leading to the almost surely consensus behavior, which partially explains the mechanism that the group with large size and strong leadership can tolerate strong noise produced by communication uncertainty. The second noise comes from environmental uncertainty, fluctuates the system and makes opinions diverse and inclusive. We derive a stochastic model with additive noise, show that the noise from the environment destroys the leadership. However, the relative fluctuations of the followers’ opinions around the leader’s opinion have a uniformly bounded variance, which means they are still in a same group with the leader’s control. Finally, numerical simulations are performed to confirm theoretical results and explore more findings.

Tuning the Noise‐Driven Magnetocaloric Effect in Doped GaAs Quantum Dot in View of Tsallis Entropy

AbstractCurrent inspection strives to perform a detailed elaboration of the Tsallis entropy‐based magnetocaloric effect (MCE) of GaAs quantum dot (QD) doped with Gaussian impurity and under applied Gaussian white noise. Noise takes additive and multiplicative pathways for its entry to the doped QD. MCE has been found to decrease following the enhancement of temperature. The enquiry manifests magnetic field‐induced confinement and consequent decline in the disorder of the system during the gradual variation of several physical parameters. However, harnessing the extent of above confinement involves extremely sensitive and delicate control over the resultant influence of the particular physical parameter undergoing change, its range of magnitude, application of noise and its mode of inclusion. Importantly, MCE has been found to maximize only during the change in the strength of the dopant potential under multiplicative noise. And, on most occasions, multiplicative noise profoundly raises the MCE over its value under the state devoid of noise and under additive noise.

Unveiling image source: Instance-level camera device linking via context-aware deep Siamese network

Unveiling the source of an image is one of the most effective ways to validate the originality, authenticity, and reliability in the field of digital forensics. Source camera device identification can identify the specific camera device used to take a photo under investigation. While great progress has been made by the photo-response non-uniformity (PRNU)-based methods over the past decade, the challenge of instance-level source camera device linking, which verifies whether two images in question were captured by the same camera device, remains significant. This challenge is mainly due to the absence of auxiliary images to construct a clean camera fingerprint for each camera, particularly dealing with small image sizes. To overcome this limitation, in this paper, we formulate the task of source device linking as a binary classification problem and propose a simple yet effective framework based on a context-aware deep Siamese network. We take advantage of a Siamese architecture to extract the intrinsic camera device-related noise patterns from a pair of image patches in parallel for comparisons without any auxiliary images. Moreover, a recurrent criss-cross group is utilized to aggregate contextual information in the noise residual maps to alleviate the problem that PRNU noise maps are easily contaminated by the additive noises from image contents. For reliable device linking, we employ a patch-selection strategy on a pair of test images to adaptively choose suitable image patch pairs according to image contents. The final decision of a pair of test images is obtained from the average similarity score of the selected image patch pairs. Compared with existing state-of-the-art methods, our proposed framework can achieve better performance on both the tasks of source camera identification and source device linking without any prior knowledge, i.e., reliable camera fingerprints, regardless of whether the camera devices are “seen” or “unseen” in the training stage. The experimental results on two standard image forensic datasets demonstrate that the proposed method not only shows robustness with respect to different image patch sizes and image quality degenerations, but also has a generalization ability across digital camera and smartphone devices.

Noise-resistant sharpness-aware minimization in deep learning

Sharpness-aware minimization (SAM) aims to enhance model generalization by minimizing the sharpness of the loss function landscape, leading to a robust model performance. To protect sensitive information and enhance privacy, prevailing approaches add noise to models. However, additive noises would inevitably degrade the generalization and robustness of the model. In this paper, we propose a noise-resistant SAM method, based on a noise-resistant parameter update rule. We analyze the convergence and noise resistance properties of the proposed method under noisy conditions. We elaborate on experimental results with several networks on various benchmark datasets to demonstrate the advantages of the proposed method with respect to model generalization and privacy protection.

The Gaussian Multiple Access Wiretap Channel With Selfish Transmitters: A Coalitional Game Theory Perspective

The Gaussian Multiple Access Wiretap Channel With Selfish Transmitters: A Coalitional Game Theory Perspective

Human foot force informs balance control strategies when standing on a narrow beam.

Despite the abundance of studies on the control of standing balance, insights about the roles of biomechanics and neural control have been limited. Previous work introduced an analysis combining the direction and orientation of foot-ground forces. The "intersection point" of the lines of actions of these forces exhibited a consistent pattern across healthy, young subjects when computed for different frequency components of the center of pressure signal. To investigate the control strategy of quiet stance, we applied this intersection point analysis to experimental data of 15 healthy, young subjects balancing in tandem stance on a narrow beam and on the ground. Data from the sagittal and frontal planes were analyzed separately. The task was modeled as a double-inverted pendulum controlled by an optimal controller with torque-actuated ankle and hip joints and additive white noise. To test our prediction that the controller that minimized overall joint effort would yield the best fit across the tested conditions and planes of analyses, experimental results were compared with simulation outcomes. The controller that minimized overall effort produced the best fit in both balance conditions and planes of analyses. For some conditions, the relative penalty on the hip and ankle joints varied in a way relevant to the balance condition or to the plane of analysis. These results suggest that unimpaired quiet balance in a challenging environment can be best described by a controller that maintains minimal effort through the adjustment of relative ankle and hip joint torques. NEW & NOTEWORTHY This study explored balance control in humans during a challenging task using the novel intersection point analysis, based on foot-ground force direction and point of application. Experimental data of subjects standing on a narrow beam in tandem stance were compared with modeling results of a double-inverted pendulum. The analysis showed that individuals minimized effort by adjusting ankle and hip torques, shedding light on the interplay of biomechanics and neural control in maintaining balance.

Frequency and damping factor estimation of damped sinusoid by using DFT and DTFT

An algorithm for estimating the frequency and damping factor of damped complex sinusoidal signal is presented. After the rough estimation of frequency by discrete Fourier transform (DFT), the fine estimation of frequency and damping factor is achieved by utilizing the DFT bin with the maximum magnitude and two arbitrarily spaced discrete time Fourier transform (DTFT) spectral lines located on the same side of the maximum DFT spectrum bin. Then the theoretical mean square error (MSE) in additive white noise is analyzed. Simulation results show that the proposed algorithm tracks the Cramer-Rao lower bound (CRLB), and the parameters estimation accuracy is higher than the competing algorithms in most situations. The computational requirements are almost the same as several existing methods.

Asymptotic behavior of non-autonomous fractional stochastic lattice FitzHugh–Nagumo system driven by linear mixed white noise

This paper is concerned with the asymptotic behavior of the non-autonomous fractional stochastic lattice FitzHugh–Nagumo system driven by the linear mixed white noise, which simultaneously contains linear additive noise and multiplicative noise. For the sake of the long-term behavior of the system we considered, we need to utilize a different Ornstein–Uhlenbeck transformation than the general one. First, the existence and uniqueness of pullback random attractors are demonstrated. Then, we prove the upper semicontinuity of random attractors when the intensity of noise approaches zero.

Accurate drift-invariant single-molecule force calibration using the Hadamard variance

Single-molecule force spectroscopy (SMFS) techniques play a pivotal role in unraveling the mechanics and conformational transitions of biological macromolecules under external forces. Among these techniques, multiplexed magnetic tweezers (MT) are particularly well suited to probe very small forces, ≤1 pN, critical for studying noncovalent interactions and regulatory conformational changes at the single-molecule level. However, to apply and measure such small forces, a reliable and accurate force-calibration procedure is crucial. Here, we introduce a new approach to calibrate MT based on thermal motion using the Hadamard variance (HV). To test our method, we perform bead-tether Brownian dynamics simulations that mimic our experimental system and compare the performance of the HV method against two established techniques: power spectral density (PSD) and Allan variance (AV) analyses. Our analysis includes an assessment of each method’s ability to mitigate common sources of additive noise, such as white and pink noise, as well as drift, which often complicate experimental data analysis. We find that the HV method exhibits overall similar or higher precision and accuracy, yielding lower force estimation errors across a wide range of signal/noise ratios (SNRs) and drift speeds compared with the PSD and AV methods. Notably, the HV method remains robust against drift, maintaining consistent uncertainty levels across the entire studied SNR and drift speed spectrum. We also explore the HV method using experimental MT data, where we find overall smaller force estimation errors compared with PSD and AV approaches. Overall, the HV method offers a robust method for achieving sub-pN resolution and precision in multiplexed MT measurements. Its potential extends to other SMFS techniques, presenting exciting opportunities for advancing our understanding of mechanosensitivity and force generation in biological systems. To make our methods widely accessible to the research community, we provide a well-documented Python implementation of the HV method as an extension to the Tweezepy package.

Prediction of approaching trains based on H‐ranks of track vibration signals

AbstractThis paper introduces a method for forecasting the arrival of trains by analyzing track vibration signals. The proposed algorithms, based on H‐ranks of track vibration signals, can generate early alerts for approaching trains. These algorithms are robust to additive noise and environmental conditions. The theoretical foundation of the method involves the application of matrix operations to detect significant changes in vibration patterns, indicating an approaching train.

Wide-sense stationary discrete-time random processes: Analyses in the discrete-Fresnel and discrete-frequency domains

This article investigates the expectation and autocorrelation functions of random processes in the discrete-Fresnel and the discrete-frequency domains, originating from wide-sense stationary (WSS) random processes in the discrete-time domain. Through mathematical deduction, it is demonstrated that WSS random processes in the discrete-time domain exhibit the same expectation and autocorrelation functions as transformed to the discrete-Fresnel domain. In contrast, this property is not observed in the discrete-frequency domain. White, pink, and brown random processes are considered to illustrate these behaviors in the discrete-Fresnel and discrete-frequency domains. Furthermore, this study numerically compares data communication systems based on single-carrier with cyclic prefix, orthogonal chirp-division multiplexing, and orthogonal frequency-division multiplexing schemes, which are respectively designed in the discrete-time, discrete-Fresnel, and discrete-frequency domains. The numerical results show that orthogonal chirp-division multiplexing and single-carrier with cyclic prefix schemes achieve comparable communication performance under any additive noise modeled by a wide-sense stationary random process, regardless of the chosen linear equalization or the absence of it.

The stochastic heat equation with fractional time and fractional time–space white noise

We give an introduction to the time-fractional stochastic heat equation driven by 1+d-parameter fractional time–space white noise, in the following two cases:(i) With additive fractional time–space white noise(ii) With multiplicative time–space Brownian noiseThe fractional time derivative is interpreted as the Caputo derivative of order α∈(0,2) and we assume that the Hurst coefficient H=(H0,H1,H2,…,Hd) of the time–space fractional white noise is in (12,1)1+d.We find an explicit expression for the unique solution in the sense of distribution of the equation in the additive noise case (i).In the multiplicative case (ii) we show that there is a unique solution in the Hida space (S)∗ of stochastic distributions and we show that the solution coincides with the solution of an associated fractional stochastic Volterra equation. Then we give an explicit expression for the solution of this Volterra equation. See Appendix below or Di Nunno et al. (2008), Holden et al. (2010) and Øksendal (2023) for more information about Hida spaces and white noise theory.A solution Y(t,x) is called mild if E[Y2(t,x)]<∞ for all t,x. For both the additive noise case and the multiplicative noise case we show that if α≥1 then the solution is mild if d=1 or d=2, while if α<1 the solution is not mild for any d.This paper is partly a survey paper, explaining the concepts and methods behind the results. It is also partly a research paper, in the sense that some results are new, to the best of our knowledge.

SITF: A Self-Supervised Iterative Training Framework for Point Cloud Denoising

Despite existing supervised point cloud denoising methods having made great progress, they require paired ideal noisy-clean datasets for training which is expensive and impractical in real-world applications. Moreover, they may perform the denoising process multiple times with fixed network parameters for better denoising results at test time. To address above issues, this paper proposes a self-supervised iterative training framework (SITF) for point cloud denoising, which only requires single noisy point clouds and a noise model. Given an off-the-shelf denoising network and original noisy point clouds, firstly, an intermediate noisier-noisy dataset is created by adding additional noises from the known noise model to noisy point clouds (i.e. learning targets). Secondly, after training on the noisier-noisy dataset, the denoising network is employed to denoise the original noisy point clouds to obtain the learning targets for the next iteration. The above two steps are iteratively and alternatively performed to get a better and better trained denoising network. Furthermore, to get better learning targets for the next round, this paper also proposes a novel iterative denoising network (IDN) architecture of stacked source attention denoising modules. The IDN explicitly models the iterative denoising process internally within a single network via reforming the given denoising network. Experimental results show that existing supervised networks trained through the SITF can achieve competitive denoising results and even outperform supervised networks under high noise conditions. The source code can be found at: https://github.com/VCG-NJUST/SITF.

A novel Deep High-level Concept-mining Jointing Hashing Model for unsupervised cross-modal retrieval

Unsupervised cross-modal hashing has achieved great success in various information retrieval applications owing to its efficient storage usage and fast retrieval speed. Recent studies have primarily focused on training the hash-encoded networks by calculating a sample-based similarity matrix to improve the retrieval performance. However, there are two issues remain to solve: (1) The current sample-based similarity matrix only considers the similarity between image-text pairs, ignoring the different information densities of each modality, which may introduce additional noise and fail to mine key information for retrieval; (2) Most existing unsupervised cross-modal hashing methods only consider alignment between different modalities, while ignoring consistency between each modality, resulting in semantic conflicts. To tackle these challenges, a novel Deep High-level Concept-mining Jointing Hashing (DHCJH) model for unsupervised cross-modal retrieval is proposed in this study. DHCJH is able to capture the essential high-level semantic information from image modalities and integrate into the text modalities to improve the accuracy of guidance information. Additionally, a new hashing loss with a regularization term is introduced to avoid the cross-modal semantic collision and false positive pairs problems. To validate the proposed method, extensive comparison experiments on benchmark datasets are conducted. Experimental findings reveal that DHCJH achieves superior performance in both accuracy and efficiency. The code of DHCJH is available at Github.

Multicarrier dual‐index permuted chaos shift keying system for efficiency improvement in chaotic communications

AbstractWe propose a multicarrier chaotic system that incorporates dual‐index and M‐ary modulation techniques. The proposed system improves the energy and spectrum efficiencies with a tradeoff in computational complexity. In addition, it applies index modulation to select the chaotic signal and permuted chaotic signals along with ‐ary modulated symbols. As a result, the spectral and energy efficiencies are increased in chaotic systems. Theoretical expressions are derived to calculate the bit error rate of the proposed system under additive white Gaussian noise and Rayleigh fading channels. The equations are validated through Monte Carlo simulations, whose results demonstrate a reduction in bit error rate for the proposed system. The data rate, spectral efficiency, energy efficiency, and computational complexity of the proposed system are analyzed and compared with those of existing multicarrier chaotic systems.

Ultrasound image smoothing based on adaptive and non-adaptive filters.

To model adaptive and non-adaptive filters to ensure smooth ultrasound images. The comparative study was conducted at Al-Yarmouk Teaching Hospital, Al Mustansiriyah University, Baghdad, Iraq, in 2019, and comprised ultrasound images of kidney (303x208 pixel) and foetus (111x109 pixel). These images were smoothed based on 8 filters; 1 non-adaptive (median), and 7 adaptive enhanced filters (Gamma, Wiener, Lee, Frost, Kuan, Adaptive Lee and Adaptive Frost). They were applied to the images by windows measuring 3x3, 5x5, 7x7. The additive noise and the multiplicative noise factor were calculated using histogram to determine the noise type for each image. Statistical criteria included mean square error, normalised absolute error and signalto- noise ratio. The relationship between noise ratio and filter type showed that Wiener was the best filter and the best sliding window was 3x3. The worst filters were Gamma, EFrost and Kuan. The relationship between sliding window size and noise ratio for all the smoothing filters clearly identified the best filter for the type of noise.

Design of optimal control strategy for range extended electric vehicles considering additional noise, vibration and harshness constraints

Extended-range electric vehicles are considered to be the most promising pure electric vehicles that can achieve lower fuel consumption and emissions. However, during vehicle operation, range extenders cause an increase in noise, vibration and sound vibration harshness (Noise, Vibration, Harshness, NVH) during startup and operation, which can significantly reduce passenger comfort. Therefore, this study proposes a multi-objective predictive energy management strategy (MPC-NVH-Rapid DP) that can find a balance between the range extender fuel consumption, power cost, and range extender NVH. First, the multi-objective energy management problem is formulated as a total operating cost minimization problem, and an optimization strategy for range extender point selection considering NVH is designed. Subsequently, the accuracies of Long Short-Term Memory (LSTM), Bidirectional LSTM (Bi-LSTM) and Convolutional Neural Network-LSTM (CNN-LSTM) for speed prediction in model predictive control (MPC) are compared, and a multi-objective algorithm based on MPC and NVH considerations is proposed as an energy management optimizer. Finally, the real-time performance of the proposed MPC-NVH-Rapid DP strategy is verified by MATLAB/Simulink and hardware-in-the-loop tests, and the results show that the proposed strategy achieves better NVH performance and improves driving comfort at the expense of partial fuel economy.

Quantum-secured LiDAR with Gaussian modulated coherent states

LiDAR systems that rely on classical signals are susceptible to intercept-and-recent spoofing attacks, where a target attempts to avoid detection. To address this vulnerability, we propose a quantum-secured LiDAR protocol that utilizes Gaussian modulated coherent states for both range determination and spoofing attack detection. By leveraging the Gaussian nature of the signals, the LiDAR system can accurately determine the range of the target through cross-correlation analysis. Additionally, by estimating the excess noise of the LiDAR system, the spoofing attack performed by the target can be detected, as it can introduce additional noise to the signals. We have developed a model for target ranging and security check, and conducted numerical simulations to evaluate the Receiver Operating Characteristic (ROC) of the LiDAR system. The results indicate that an intercept-and-recent spoofing attack can be detected with a high probability at a low false-alarm rate. Furthermore, the proposed method can be implemented using currently available technology, highlighting its feasibility and practicality in real-world applications.