Input Noise Research Articles

Random Number Generator

Illustration plays an important role during education: The Galton board is a suitable tool for illustrating random processes and explaining probability distributions. We have created this tool in a virtual version, which facilitates data collection for statistical processing of experimental data and also enables the study of non-symmetrical distributions. The random processes on the device are simulated, which requires a random number generator. Since there were some doubts about the software-generated pseudo-random numbers, we created a true random number generator based on the input noise of microcontrollers.

Forward dynamics and memory effect in a fractional order chemostat minimal model with non-monotonic growth

This paper aims to study the long-time dynamics of chemostat model with non-monotonic growth, subject to random bounded disturbances on the input flow. To be precise, we first prove existence and uniqueness of global positive solution of such model and then construct the compact forward absorbing and attracting sets, which are independent of the realizations of the input noise. Moreover, we prove the conditions for biomass extinction and species persistence. In particular, fractional operator has a noticeable effect, in such a way that it can delay the rate of extinction and persistence and amplitude of oscillator. Numerical simulations are presented to verify the theoretical results.

Adaptive denoising combined model with SDAE for transient stability assessment

• A targeted denoising model (TDM) is constructed to obtain the satisfactory results when the data with similar degree of noise input. • A noise recognition model is constructed to output a set of weight values which can reflect the distribution of the noise in the input data. • An ADCM composed of multiple strong classifiers TDMs and the noise recognition model is proposed. ADCM can adaptively output the results with high confidence when the real data with different noise interference are input. Transient stability assessment (TSA) is usually considered as a challenging problem in power systems. During the measurement and transmission of real-time data in power grids, different degrees of noise may occur. The real noise is complex and hard to predict, so the current methods cannot output satisfied TSA results. Therefore, an adaptive denoising combined model (ADCM) is proposed in this paper. The ADCM can adaptively output the results of TSA in complex noise environment. Firstly, the stacked denoising auto-encoder based model trained by data with expected noise is proposed, which is called targeted denoising model (TDM). The TDM can output satisfied results for specific noisy data. Then, the ADCM composed of multiple TDMs and a noise recognition model is constructed. The recognition model is a regression model based on deep neural networks. When actual data are input, the recognition model can output weight values for TDMs. And when the actual noise is similar to the noise trained for the TDM, the weight obtained is great for this model. The final results are obtained by assigning the weights to the results of each TDM. The effectiveness of this method is verified by simulation results in IEEE 39-bus system and realistic system.

Investigating system performance in 16 channel WDM network at minimum attenuation window for pre-defined input noise level

Abstract Input and output optical signal-to-noise ratio in 16-channel WDM network are investigated for predefined input noise level when design is set at minimum attenuation window. By keeping the input OSNR constant, gain and noise figure are computed along with SNR at both transmitting and receiving terminals for two different data rates and also with short-to-long range of trans-receiver distance. Corresponding variations of all the network properties, i.e., maximum Q-factor, eye height and noise figure are measured for estimation of performance degradation. Optimization in terms of both bit rate and channel distance is required, as evident from simulation result, for best possible system performance. Setting up of input noise level at 20 dBm indicates the fact of real-time compatibility with optimum value of output OSNR is revealed at a particular distance, which suggests the best result for communication system design.

Absolute 3D Human Pose Estimation Using Noise-Aware Radial Distance Predictions

We present a simple yet effective pipeline for absolute three-dimensional (3D) human pose estimation from two-dimensional (2D) joint keypoints, namely, the 2D-to-3D human pose lifting problem. Our method comprises two simple baseline networks, a 3D conversion function, and a correction network. The former two networks predict the root distance and the root-relative joint distance simultaneously. Given the input and predicted distances, the 3D conversion function recovers the absolute 3D pose, and the correction network reduces 3D pose noise caused by input uncertainties. Furthermore, to cope with input noise implicitly, we adopt a Siamese architecture that enforces the consistency of features between two training inputs, i.e., ground truth 2D joint keypoints and detected 2D joint keypoints. Finally, we experimentally validate the advantages of the proposed method and demonstrate its competitive performance over state-of-the-art absolute 2D-to-3D pose-lifting methods.

IMAGE ORIENTATION BY EMBEDDING IN A GAN LATENT SPACE

Abstract. Estimation of an image exterior orientation is required in multiple tasks including 3D reconstruction, autonomous driving and navigation, and single-photo 3D reconstruction. The problem can be easily solved if some reference points (keypoints) with known coordinates in the reference frame are detected in the image. While multiple robust keypoint detectors were developed, estimation of an image exterior orientation from a single image remains challenging in many cases. For example, repeating structures in the scene or absence of textures can reduce the performance of keypoint detectors. In this paper, we propose an algorithm for estimation of an image exterior orientation that leverages the latent space of Generative Adversarial Network (GAN). We propose a modification of the StyleGAN2 model that we term ExteriorGAN. Unlike the StyleGAN2 that generates random images from a random noise vector z, we aim training a mapping from a random vector z and a given image exterior orientation p G : (z,p) → A. Our model generates random images for a constant exterior orientation p and random z that have constant geometry but differs in the scene appearance (e.g. different light direction or intensity). We perform embedding of the image A into the latent space w and reconstruct the input noise vector z and exterior orientation parameters w using a stochastic gradient descent. We developed a dedicated dataset with 50k images and corresponding orientation parameters to train and validate our ExteriorGAN model. The results of evaluation demonstrate that our algorithm allows estimation of the exterior orientation of an image with respect to a known 3D scene. The accuracy of the exterior orientation is comparable with modern state-of-the-art methods. The camera pose can be recovered with a mean error of 50 mm for a working space of 5 by 5 meters.

Adaptive Weiner filtering with AR-GWO based optimized fuzzy wavelet neural network for enhanced speech enhancement.

Speech signal enhancement is a subject of study in which a large number of researchers are working to improve the quality and perceptibility of speech signals. In the existing Kalman Filter method, the short-time magnitude or power spectrum due to random variations of noise was a serious problem and the signal-to-noise ratio was very low. This issue severely reduced the perceived qualityand intelligibility of enhanced speech. Thus, this paper intent to develop an improved speech enhancement model and it includes"training phase and testing phase". In the training phase, the input noise corrupted signal is initially fed as input to both STFT-based noise estimation and NMF-based spectrum estimation forestimating the noise spectrum and signal spectrum, respectively. The obtained noise spectrum and the signal spectrum are fed as input to the Wiener filter and these filtered signals are subjected to Empirical Mean Decomposition (EMD).Since, tuning factor η plays a key role in Wiener filter, it has to be determined for each signal and from the denoised signal the bark frequency is evaluated. The computed bark frequency is fed as input to the learning algorithm referred as Fuzzy Wavelet Neural Network (FW-NN)for detecting the suited tuning factor η for the entire input signal in Wiener filter.An Adaptive Randomized Grey Wolf Optimization (AR-GWO) is proposed for proper tuning of the tuning factor η referred as tuned tuning factor (η tuned ). The proposed AR-GWO is the improved version of standard Grey wolf optimization (GWO). In the testing phase, the training is accomplished initially and from which the tuning factor is gathered for each of the relevant input signal. Then, the properly tuned tuning factor (η tuned ) from FW-NN is fed as input to EMD via adaptive wiener filter for decomposing the spectral signal and the output of EMD is denoised enhanced speech signal. At last, the performance of the adopted approach is evaluated to the existing approaches in terms of various metrics. In particular, the computation time of the adopted AR-GWO model is 34.07%, 43.57%, 28.86%, 38.88%, and 16.03% better than the existing GA, ABC, PSO, FF, and GWO approaches respectively.

A Variable Step Size Gradient-Descent TLS Algorithm for Efficient DOA Estimation

The direction-of-arrival (DOA) estimation model of this brief is based on deviation compensation model with input noise, and the performance of the traditional Least mean square (LMS) adaptive algorithm shows poor performance. Instead, total least squares (TLS) algorithm is widely used in models which contains input noise. Therefore, we present TLS algorithm for DOA estimation, which is used to update weight coefficient by searching the peak of the spatial spectrum to estimate the direction of the angle of arrival. Due to the unsatisfactory DOA estimation performance on fixed step algorithms, a variable step-size gradient descent total least-squares (VSS-GDTLS) is proposed. The variable step size strategy is derived by applying the instantaneous augmented weight vector and the estimated signal power. Moreover, the convergence of the proposed algorithm is analyzed. Finally, simulation results show the superiority of the VSS-GDTLS algorithm than GDTLS and the other adaptive algorithms.

Circuit Design for a 4-Channel EEG Pre-processing System with Programmable Output Control Capability

The design of a 4-channel Electro-encephalography (EEG) pre-processing circuit using 0.6m CMOS technology is proposed in this paper. The circuit is divided into two parts. To amplify and differentiate the Alpha, Beta, Delta, and Theta waves from the original EEG signal, a mixed-mode EEG pre-processing system with high gain, low power, and low input noise amplifier and four Gm-C band-pass filters driven by a 1V single power supply are designed. The 4-channel pre-processing system is implemented by connecting four of them in parallel. Second, a digital control circuit consisting of a Serial Input Parallel Output (SIPO) Shift Register and SRAM array is constructed and used to control the output of the pre-processing system, which is powered by a 3.8V power supply.

Design of chopper operational amplifier for leakage current protector chip

Due to the low frequency and small amplitude of the input signal of the leakage current protection chip, an operational amplifier with low offset and low power consumption is designed by using the chopper stabilization technique and the dynamic element matching technique. Based on the TSMC 0.18um process, the spectre environment in cadence is used for design and simulation. The power supply voltage is 1.8V. The post simulation results show that under the TT corner, the average current of the operational amplifier is 139uA, the loop gain is 85.9dB, the phase margin is 68.01°, and the power supply rejection ratio(PSRR) is 101.62dB@50Hz, the common mode rejection ration(CMRR) is 104.52dB@50Hz, the equivalent input noise is 129.86nv/sqrt(Hz)@50Hz, and the value of equivalent input offset voltage is 8.2uV(mean).

A new development algorithm for permeability prediction: A new milestone

Permeability is one of the most important reservoir rock parameters in petroleum engineering, reservoir, and exploitation. This parameter causes the movement of hydrocarbon reserves in the reservoir rock. Therefore, it is an important parameter from the economic point of view because it greatly impacts the amount of extraction from the reservoir rock. In this study, the combined RBFNN-GA algorithm and 200 data sets collected from a field in the Middle East were used to predict permeability. Water saturation, porosity, and specific surface are the input variables used in this study. GA has advantages such as solving complex optimization problems of continuous functions and multi-objective problems. The advantages of RBF neural networks are that they are easy to design, strongly tolerant to input noise, and have good generalization. The RBFNN-GA model has the advantages of both algorithms. RBFNN-GA algorithm and experimental models have been compared in terms of performance accuracy. The results show that RBFNN-GA with STD = 89.8 and R-square = 0.9011 for the total data set obtained from a field in the Middle East has better accuracy and performance in predicting permeability than experimental models. Compared to other neural network methods, the RBFNN-GA model has a higher performance accuracy and is efficient for predicting other parameters. Oil researchers and engineers can use this method to predict other parameters in their studies and research.

SSP-WGAN-Based Data Enhancement and Prediction Method for Cement Clinker f-CaO

Aiming at the problem of low prediction accuracy of traditional prediction models due to the limited labeled sample data and the imbalance of multitimescale sample data in the cement production process, a cement clinker-free calcium oxide (f-CaO) data enhancement and prediction model based on semisupervised prediction Wasserstein generative adversarial networks (SSP-WGANs) is proposed in this article. The model is constructed by WGAN and a prediction model. In the generator of WGAN, the traditional noise input is replaced by time-series matrices composed of related variables affecting the f-CaO content of cement clinker. The generator maps the input high-dimensional time-series data into low-dimensional labeled values through its internal convolutional layer, which can fill in the missing labeled values of the input unlabeled samples. The generated labels are spliced with the related variables affecting the f-CaO content of clinker according to the timescale relationship and are uniformly sent to the discriminator with the real sample pairs for discrimination, thus eliminating the influence of multiple timescales. The generated labeled data are matched with the unlabeled data and used to expand the training set of the prediction model composed of CNN-GRU, which significantly improves the prediction accuracy of the model. The results show that the SSP-WGAN model with data enhancement has higher accuracy and stronger robustness.

C-Band GaN Dual-Feedback Low-Noise Amplifier MMIC with High-Input Power Robustness

In this paper, using the 0.2 μm ETRI GaN HEMT process, we developed a C-band GaN dual-feedback low-noise amplifier MMIC for an RF receiver module that requires high-input power robustness. By applying a feedback microstrip line at the source of the transistor and series resistor-capacitor (RC) feedback between the gate and the drain of the transistor, we obtained stable amplifier operation and a compromised impedance trace for both input impedance matching and noise matching while suppressing performance degradation of the maximum available gain and minimum noise figure. The developed low-noise amplifier MMIC, which implements simple matching circuits by using biasing elements as matching elements, had a linear gain of more than 21.4 dB and a noise figure of less than 1.91 dB in the wide bandwidth of 4.3–7.4 GHz. Under the single-tone power test, the low-noise amplifier MMIC had an output P1dB of 14.3–20.1 dBm, and the two-tone intermodulation distortion measurement exhibited an input third-order intercept point (IIP3) of 2.2–5.6 dBm in the same frequency range as the above.

Multi-photon-addition amplified coherent state

State [Formula: see text] and state [Formula: see text] are same to state [Formula: see text], which is called as multi-photon-addition amplified coherent state (MPAACS) by us. Here, [Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text], and [Formula: see text] are photon number operator, creation operator, coherent state, gain factor, and an integer, respectively. We study mathematical and physical properties for these MPAACSs, including normalization, photon component analysis, Wigner function, effective gain, quadrature squeezing, and equivalent input noise. Actually, the MPAACS, which contains more nonclassicality, is an amplified version of photon-added coherent state (PACS) introduced by Agarwal and Tara [Phys. Rev. A 43, 492 (1991)]. Our work provides theoretical references for implementing amplifiers for light fields.

Experimental Evaluation of Display Field Communication Based on Machine Learning and Modem Design

Display field communication (DFC) is a frequency-domain unobtrusive display-to-camera (D2C) communication, in which an electronic display serves as a transmitter and a camera serves as a receiver. In this paper, we propose a machine learning-based DFC scheme and evaluate its performance in a lab test scenario. First of all, we adopt the Discrete Cosine Transform (DCT) to transform a spatial-domain image into its spectral-domain equivalent. To reduce the computational complexity during the data-embedding process, addition allocation and subtraction data retrieval techniques are used. Moreover, channel coding is applied to overcome the data error caused by the optical wireless channel. In particular, robust turbo coding is used for error detection and correction. Afterward, we perform the experiments to validate the performance of the proposed system. After capturing the displayed image with a camera, data restoration is done using a deep learning technique. Extensive real-world experiments were performed considering various geometric distortions, noise, and different standard input images. As a result, we found that by increasing the transmit display image size (upsampling), the overall error rate can be reduced. In addition, real-world noise analysis is performed and it is notified that the actual noise is dominant in the low-frequency region of an image. The experimental results confirm the robust performance of the proposed DFC scheme and show that an error-free performance can be achieved up to a distance of 1m in the given lab test environment setting.

A Hybrid Image Denoising Method Based on Discrete Wavelet Transformation with Pre-Gaussian Filtering

Background/Objectives: At the time of acquisition and transmission noise is embedded with the images. It introduces new but unwanted information (noise) in images. The elimination of noise to analyze such data is an essential step in preprocessing. The purpose of this study is to propose a novel image denoising approach to recover original images at high noise densities without introducing unwanted artifacts. Methods: A new hybrid method based on approximation subband thresholding with pre-Gaussian filtering is presented in this study. Google Colab as a platform and python as a programming language is used for the implementation of the proposed technique. To evaluate the performance Peak Signal to Noise Ratio (PSNR) is chosen. The standard jpeg images (Cameraman, Lena, Astronaut, Cat) have been taken as an input and random noise with different noise ratios (s =0.05,0.20,0.30,0.50) is applied to get the noisy images for the experiment. In random noise scenarios, the proposed method experimented on different grayscale standard images, and performance is compared with different existing methods. Findings: The standard images with different noise ratios are denoised by the proposed method, and the quality of images is calculated in terms of PSNR. The results obtained from the proposed method on different standard images improve PSNR (PSNR= 25.80dB, s =0.50) at high noise levels significantly. Novelty: Gaussian filter improve the quality of images. However, when wavelet decomposition is blended with filtered image and thresholding is applied on approximation band improved the quality of images. Hence, the proposed method has a wide area of application to improve image quality in the field of character recognition, agriculture, medical science, and remote sensing. Keywords: Gaussian Filter; Discrete Wavelet Thresholding; Image denoising; Image Processing

Unambiguous Direction Estimation and Localization of Two Unresolved Targets via Monopulse Radar

Traditional monopulse radar cannot resolve two closely spaced targets present in one resolution cell (range and Doppler) by means of the monopulse ratio. This study presents a closed-form solution to resolve the directions of arrival of two unresolved targets using a single snapshot of four independent channels in phase comparison monopulse radar. If both targets have the same elevation or same azimuth direction, the proposed scheme cannot estimate their directions. To estimate the direction of such targets, an extra antenna is required. The impact of input noise power, the targets’ direction, and phase difference of the targets’ signal on the accuracy of angle estimation are also explained. The numerical simulation result validates the effectiveness of the presented scheme.

A 23–31 GHz high‐gain GaN low‐noise amplifier using T‐type matching networks and multiple feedback techniques for broadband 5G millimeter‐wave applications

AbstractA 23–31 GHz three‐stage low‐noise amplifier (LNA) based on 100 nm gallium nitride‐on‐Si technology is presented with a total circuit size of 2.0 × 1.4 mm2. To achieve an excellent noise figure (NF) and gain, a special inductive dual‐source feedback network is used; it contributes to noise matching and input conjugate matching by resonating with the parasitic gate‐source capacitance of the transistor. T‐type networks with low‐quality factor are carefully employed for broadband matching circuits to extend the bandwidth. Meanwhile, a lossy RLC feedback network of the output stage can further obtain a wider frequency band and improve the stability through Miller approximation analysis of the input quality factor and gain of the closed‐loop system. The proposed LNA achieves a NF of 1.9–2.5 dB, a small signal gain of 24–27 dB, and an output 1 dB compression point (P1dB) of 16 dBm across the band.

A Low-power Neural Signal Acquisition Analog Front-end IC for Closed-loop Neural Interfaces

This paper presents an ultra-low-power neural signal recording analog front-end IC (AFE) which is comprised of a low-noise amplifier (LNA) followed by a programmable gain amplifier (PGA). To achieve good overall performance while handling stimulation artifacts in closed-loop neural interface systems, the proposed LNA is designed for moderate gain, low input noise, and, good linearity while the PGA is included for additional gain and tunable bandwidth functions. Implemented using 0.18-㎛ CMOS process, the AFE achieves a measured closed-loop gain of 20 to 40 ㏈ and integrated input referred noise of 3.39 ㎶rms over 1 ㎐ to 6.5 ㎑. The IC consumes 1.49 ㎼ at 1-V supply and the noise efficiency factor is 1.93. For less than 1% total harmonic distortion, the AFE can accommodate up to 60 ㎷pp and 280 ㎷pp of differential and common-mode stimulation artifacts, respectively.

EA-CNN: A smart indoor 3D positioning scheme based on Wi-Fi fingerprinting and deep learning

Accurate indoor location information in multi-building/floor environments is essential for establishing many indoor location-based services (LBS). Wi-Fi fingerprinting with received signal strength indicators (RSSI) has become one of the most practical techniques to localize users indoors. However, the fluctuation of wireless signals caused by fading, the multipath effect, and device heterogeneity leads to considerable variations in RSSIs, which poses a challenge to accurate localization. This paper proposes an indoor positioning method, which is constructed based on a convolutional neural network (CNN) framework. Specifically, a novel model by combining an extreme learning machine autoencoder (ELM-AE) with a two-dimensional CNN is proposed. The ELM-AE extracts critical features by reducing input dimensions, while the CNN is trained to effectively achieve significant performance in the positioning phase. To increase positioning accuracy and deal with data shortages, a data augmentation strategy by frequently adding noisy data to the original fingerprint map is devised. The statistical properties namely, the RSS values of each MAC address, are utilized to adjust input noise. The performance of the proposed system is evaluated on two Tampere and UJIIndoorLoc datasets. The experimental results show that EA-CNN achieves better performance than CNN by decreasing average positioning error up to 40.95% and 43.74% in Tampere and UJIIndoorLoc datasets, respectively. Compared to state-of-the-art deep learning-based methods, the positioning performance improves up to 68.36% in Tampere and 67.56% in the UJIIndoorLoc dataset by exploiting only 25% of the training samples.Compared to several state-of-the-art deep learning models, EA-CNN achieves higher accuracy in both positioning and floor estimation.