Data Decorrelation Research Articles

Methodology for diagnosing the technical condition of aviation gas turbine engines using recurrent neural networks (RNN) and long short-term memory networks (LSTM)

This study presents a method for diagnosing the technical condition of aviation gas turbine engines (GTE) using recurrent neural networks (RNN) and long short-term memory networks (LSTM). The primary focus is on comparing the effectiveness of these models for forecasting key operating parameters of GTEs, such as vibrations, turbine-inlet temperatures, and rotor speeds of low and high pressure. The research involved thorough data cleaning and normalization, including handling missing values, normalization using Min-Max Scaling, outlier removal, data decorrelation, and time series smoothing. The RNN and LSTM models were trained using the backpropagation through time (BPTT) algorithm to accurately forecast GTE operating parameters. The results show that both models demonstrate high forecasting accuracy, but the RNN models perform better in most parameters. For vibration parameters (VIB_N1FNT1, VIB_N1FNT2, VIB_N2FNT1, and VIB_N2FNT2), RNN models achieved lower RMSE and MAE values, confirming their higher accuracy. For temperature parameters (EGT1 and EGT2), RNN models also showed higher accuracy rates. Meanwhile, LSTM models achieved better results for some rotor speed parameters (N21 and N22). The findings emphasize the necessity of choosing the appropriate model based on the nature of data and the specifics of the parameters to be forecast. Future research may focus on developing hybrid approaches that combine the advantages of both models to achieve optimal results in diagnosing the technical condition of GTEs.

SEMbap: Bow-free covariance search and data de-correlation.

Large-scale studies of gene expression are commonly influenced by biological and technical sources of expression variation, including batch effects, sample characteristics, and environmental impacts. Learning the causal relationships between observable variables may be challenging in the presence of unobserved confounders. Furthermore, many high-dimensional regression techniques may perform worse. In fact, controlling for unobserved confounding variables is essential, and many deconfounding methods have been suggested for application in a variety of situations. The main contribution of this article is the development of a two-stage deconfounding procedure based on Bow-free Acyclic Paths (BAP) search developed into the framework of Structural Equation Models (SEM), called SEMbap(). In the first stage, an exhaustive search of missing edges with significant covariance is performed via Shipley d-separation tests; then, in the second stage, a Constrained Gaussian Graphical Model (CGGM) is fitted or a low dimensional representation of bow-free edges structure is obtained via Graph Laplacian Principal Component Analysis (gLPCA). We compare four popular deconfounding methods to BAP search approach with applications on simulated and observed expression data. In the former, different structures of the hidden covariance matrix have been replicated. Compared to existing methods, BAP search algorithm is able to correctly identify hidden confounding whilst controlling false positive rate and achieving good fitting and perturbation metrics.

Monitoring of high-dimensional and high-frequency data streams: A nonparametric approach

ABSTRACT High-dimensional and high-frequency data streams (HHDS) have become widely available with the prevalence of data-acquisition technologies. Because HHDS have the characteristics of high dimensionality and temporal correlation, traditional statistical process control (SPC) techniques cannot be used directly to address the problem of monitoring HHDS. Most extant monitoring methods commonly assume that the process observations are independent or can be described by some parametric models and face the challenge posed by high-dimensional data. Little research has been conducted on the robust monitoring of HHDS that can simultaneously accommodate high dimensions and high frequencies. Therefore, in this paper, we propose a novel nonparametric approach for monitoring HHDS, based on data decorrelation, dimension reduction, and SPC. Specifically, process observations are first sequentially decorrelated and standardized using Cholesky decomposition, and then the k -nearest neighbors classification algorithm is applied to transform uncorrelated high-dimensional data into one-dimensional data. Finally, the traditional cumulative-sum procedure is used for online monitoring, based on an empirical log-likelihood ratio test. Numerical studies have shown that our monitoring scheme is sufficiently reliable and efficient for the online monitoring of HHDS. An illustrative example about the Dow Jones 30 industrial stock prices is presented to further validate the proposed approach.

Improved time-variable transformer-based fault diagnosis method for satellite attitude control system

This study proposes a fault diagnosis method based on a time-variable transformer network for satellite attitude control system telemetry data, which has the properties of a time series and strong multivariate correlation. First, a time-attention module was utilised to determine the correlation between the data at a specific time and before that time to effectively identify the dynamic properties of the data. A variable attention module is introduced to capture the degree of correlation between various variables to achieve multivariate decoupling. Then, the interactive attention layer combines the dynamic properties of the data with the correlations between the variables to achieve dynamic data decorrelation. Finally, a linear layer is used to implement the fault diagnosis.

ОСОБЛИВОСТІ СТИСНЕННЯ МУЛЬТИСПЕКТРАЛЬНИХ ЗОБРАЖЕНЬ ІЗ ВТРАТАМИ

This paper concerns some aspects of lossy compression applied to six-component Landsat multispectral images where all component images have identical spatial resolution. It is shown that most component images are correlated although the cross-correlation coefficients for component images vary in rather wide limits – from 0.11 in one dataset and 0.72 in another dataset to almost unity in both cases. Compression based on discrete cosine transform is considered where 2D blocks have the size of 32×32 pixels. Embedded deblocking after decompression is applied as well. Lossy compression can be applied component-wise and in a 3D manner for spectral decorrelation of data where 6-point discrete cosine transform is applied. It is shown that, in the latter case, a significantly larger compression ratio can be achieved. The attained benefit in CR is about 30-80% compared to the component-wise compression or, alternatively, the benefit in peak signal-to-noise ratio can reach a few dB for the same compression ratio. The coders are studied for peak signal-to-noise ratio varying in the limits from about 23 dB to about 46 dB that correspond to image quality starting from annoying to practically invisible distortions. In addition, there are some image pre-processing operations that are able to further improve the compression ratio. They are image normalization and band re-ordering. All possible variants of band ordering are studied for both test images. The impact of band ordering on compressed image quality is analyzed in terms of rate/distortion curves for the 2D and 3D versions of the considered coder. It is demonstrated that, due to optimal ordering, a few % improvement of CR can be gained. However, the optimal band order for the considered multichannel images is not the same and this can cause problems in practice to be studied in the future.

Oil Formation Volume Factor Prediction Using Artificial Neural Network: A Case Study of Niger Delta Crudes

Artificial intelligence techniques provide an alternative to conventional empirical correlation methods when experimentally determined oil formation volume factors (OFVF) are lacking. A new mathematical model is proposed using an artificial neural network (ANN) for estimating the OFVF for the Niger Delta crude oils. The method consists of two stages: data decorrelation through principal component analysis (PCA) and OFVF estimation through ANN. Data decorrelation was used to reduce redundancy in the data which decreased the number of neurons in the hidden layer needed for an ANN to achieve high accuracy. In the development of the model, 316 data points were obtained from the Niger Delta region of Nigeria. Application of data cleaning, outliers’ elimination and PCA analysis reduced the data to 243 points. 213 data points were used to develop the model of which 75% was used for training, 15% for validation and 10% for testing. The remaining 30 data points were used to test the predictive capability of the proposed model. The results obtained were compared with widely accepted empirical correlations of Standing, Glaso, Vazquez, Ikiensikimama & Ajienka, and Al-Marhoun. The proposed new model performed better than all of them in terms of coefficient of correlation, AAPE and RMSE. Hence the ANN model will reduce cost, save time, and also predict the OFVF of Niger Delta crudes with higher precision.

A General Framework for Robust Monitoring of Multivariate Correlated Processes

Statistical process control (SPC) charts provide an important analytic tool for online monitoring of sequential processes. Conventional SPC charts are designed for cases when in-control (IC) process observations are independent and identically distributed at different observation times and the IC process distribution belongs to a parametric (e.g., normal) family. In practice, however, these model assumptions are rarely valid. To address this issue, there have been some existing discussions in the SPC literature for handling cases when the IC process distribution cannot be described well by a parametric form, and some nonparametric SPC charts have been developed based on data ranking and/or data categorization. However, both data ranking and data categorization would lose information in the original process observations. Consequently, the effectiveness of the nonparametric SPC charts would be compromised. In this article, we make another research effort to handle this problem by developing a general process monitoring framework that is robust to the IC process distribution and short-ranged serial correlation. The new method tries to preserve as much information in the original process observations as possible. Instead of using data ranking and/or data categorization, it is based on intensive data pre-processing, including data decorrelation, data transformation, and data integration. Because the distribution of the pre-processed data can be approximated well by a parametric distribution, the design and implementation of the new method is relatively simple. Numerical studies show that it is indeed robust to the IC process distribution and effective for online monitoring of multivariate processes with short-ranged serial correlation.

Methodology for Complete Decorrelation of Power Supply EM Side-Channel Signal and Sensitive Data

Electro-magnetic (EM) side channel attacks have become a serious threat to security of Internet-of-Things (IoT) devices. Power supply generated by voltage regulators is one of the most common attack targets due to its strong EM emanations. In this brief we derive analytical conditions for complete theoretical decorrelation of the power supply EM side-channel signal and the sensitive data. The output of the power supply converter is modelled as amplitude modulation (AM) of the load signal by the converter capacitance that acts as a carrier. By applying Price theorem (Papoulis and Pillai, 2002), we obtain the exact theoretical conditions that converter capacitance needs to fulfil in order to prevent EM side-channel attacks. The conditions are further adapted for practical implementation. When the proposed methodology is applied to AES measured traces, the correlation coefficient between the leaked signal and the sensitive data is 0.05. Such low correlation indicates the proposed methodology is a promising candidate against the attacks that exploit AM signals to extract sensitive data, such as, TEMPEST and active EM attacks. Test Vector Leakage Assessment (TVLA) <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\rho $ </tex-math></inline-formula> -test detects no leaky points, thereby confirming circuit protection against differential and correlation EM attacks as well.

Cyclic Nonlinear Correlation Analysis for Time Series

Principal component analysis (PCA) and kernel PCA allow the decorrelation of data with respect to a basis that is found via variance maximization. However, these techniques are based on pointwise correlations. Especially in the context of time series analysis this is not optimal. We present a novel generalization of PCA that allows to imprint any desired correlation pattern. Thus the proposed method can be used to incorporate previously known statistical dependencies between input variables into the model which is increasing the overall performance. This is achieved by generalizing the projection onto the direction of maximum variance—as known from PCA—to a projection onto a multi-dimensional subspace. We focus on the use of cyclic correlation patterns, which is especially of interest in the domain of time series analysis. Beneath introducing the presented variation of PCA, we discuss the role of this method with respect to other well-known time series analysis techniques.

Data-independent low-complexity KLT approximations for image and video coding

The Karhunen–Loève transform (KLT) is often used for data decorrelation and dimensionality reduction. The KLT is able to optimally retain the signal energy in only few transform components, being mathematically suitable for image and video compression. However, in practice, because of its high computational cost and dependence on the input signal, its application in real-time scenarios is precluded. This work proposes low-computational cost approximations for the KLT. We focus on the blocklengths N∈{4,8,16,32} because they are widely employed in image and video coding standards such as JPEG and high efficiency video coding (HEVC). Extensive computational experiments demonstrate the suitability of the proposed low-complexity transforms for image and video compression.

Low-complexity rounded KLT approximation for image compression

The Karhunen-Lo\`eve transform (KLT) is often used for data decorrelation and dimensionality reduction. Because its computation depends on the matrix of covariances of the input signal, the use of the KLT in real-time applications is severely constrained by the difficulty in developing fast algorithms to implement it. In this context, this paper proposes a new class of low-complexity transforms that are obtained through the application of the round function to the elements of the KLT matrix. The proposed transforms are evaluated considering figures of merit that measure the coding power and distance of the proposed approximations to the exact KLT and are also explored in image compression experiments. Fast algorithms are introduced for the proposed approximate transforms. It was shown that the proposed transforms perform well in image compression and require a low implementation cost.

Aerodynamic Parameter Estimation Using Reconstructed Turbulence Measurements

A classical flow-vane approach for reconstructing translational gust velocities from air data and inertial sensor measurements was improved and implemented for real-time computation. The reconstructed turbulence measurements were then included in a system identification analysis to estimate nondimensional stability and control derivatives in a longitudinal short period model using the maximum likelihood equation-error method in the frequency domain with Fourier-transform data. Flight-test results using 44 maneuvers from a subscale transport-type airplane showed that the power spectra for the reconstructed vertical gusts resembled standard Dryden or von Kármán turbulence models. Flight data in moderate and severe turbulence exhibited a decorrelation of the modeling data. In particular, pitch rate and angle-of-attack rate derivatives could both be identified from flight data about straight and level flight without special maneuvers or prior information.

Constraints on the location, depth and yield of the 2017 September 3 North Korean nuclear test from InSAR measurements and modelling

SUMMARY The Democratic People's Republic of Korea (North Korea) conducted its sixth and largest affirmed underground nuclear test on 2017 September 3. Analysis of Interferometric Synthetic Aperture Radar (InSAR) data revealed detailed surface displacements associated with the nuclear explosion. The nuclear explosion produced large-scale surface deformation causing decorrelation of the InSAR data directly above the test site, Mt. Mantap, while the flanks of the Mountain experienced displacements up to 0.5 m along the Line-of-Sight of the Satellite. We determined source parameters of the explosion using the Bayesian inversion of the InSAR data. The explosive yield was estimated as 245–271 kiloton (kt) of TNT, while the previous yield estimations range from 70–400 kt. We determined the nuclear source at a depth of 542 ± 30 m below Mt. Mantap (129.0769°E, 41.0324°N). We demonstrated that the Bayesian modelling of the InSAR data reduces the uncertainties in the source parameters of the nuclear test, particularly the yield and source depth that are otherwise poorly resolved in seismic methods.

Low-complexity 8-point DCT approximation based on angle similarity for image and video coding

The principal component analysis (PCA) is widely used for data decorrelation and dimensionality reduction. However, the use of PCA may be impractical in real-time applications, or in situations were energy and computing constraints are severe. In this context, the discrete cosine transform (DCT) becomes a low-cost alternative to data decorrelation. This paper presents a method to derive computationally efficient approximations to the DCT. The proposed method aims at the minimization of the angle between the rows of the exact DCT matrix and the rows of the approximated transformation matrix. The resulting transformations matrices are orthogonal and have extremely low arithmetic complexity. Considering popular performance measures, one of the proposed transformation matrices outperforms the best competitors in both matrix error and coding capabilities. Practical applications in image and video coding demonstrate the relevance of the proposed transformation. In fact, we show that the proposed approximate DCT can outperform the exact DCT for image encoding under certain compression ratios. The proposed transform and its direct competitors are also physically realized as digital prototype circuits using FPGA technology.

Modeling Interferometric SAR Features of Forest Canopies Over Mountainous Area at Landscape Scales

High-quality interferometric synthetic aperture radar (InSAR) data have become more available in recent years with InSAR missions such as the TanDEM-X. Theoretical coherent backscattering models of forest canopies at landscape scales are necessary for understanding the relationship between positions of scattering phase center extracted from InSAR data and spatial structures of forest stands growing on complex mountainous terrain. Unlike most existing scattering models of forest canopies that focus on the prediction of the scattering behavior within a pixel, a new model, referred to as LandSAR, was developed to fully account for compound effects of forest spatial structure, terrain, and geometrical distortion caused by slant range imaging on the scattering phase center. The LandSAR model was validated over a mountainous forest scene (about 8.8 km by 9.5 km) imaged by an airborne laser scanner. The interferogram, flattened interferogram, coherence, unwrapped phase, and digital surface model were successfully extracted from simulated InSAR data that have been processed as real InSAR data. The effects of wavelength, baseline length, land cover, and terrain features on decorrelation of simulated InSAR data were consistent with theoretical expectations. Both the height of the scattering phase center and the penetration depth were strongly correlated with forest heights. These results demonstrated that the LandSAR successfully modeled the InSAR features of forest canopies over a mountainous area at landscape scales.

Correlation in the application of the triple-to-double coincidence ratio method with unequal photomultiplier tube efficiencies

Counting data from the triple-to-double coincidence ratio (TDCR) method are by their nature highly correlated. In the general case of unequal phototube detection efficiencies, the highly correlated nature of the data combined with the need to apply methods to solve a system of equations makes proper uncertainty analysis difficult. Removing the correlations from the data prior to analysis may assist in this endeavor. This paper describes the de-correlation of TDCR data by the application of a particular linear transform, the Mahalanobis transform. A programming example in MATLAB is given and a practical example of the use of this technique in the analysis of TDCR data is presented.

Register Pre-Allocation based Folded Discrete Tchebichef Transformation Technique for Image Compression

Recently, the large size data, power and real-time processing abilities are major issues in Digital Signal Processing/multimedia applications which require an adaptable architecture. The tool used for computing data decorrelation in the image processing applications refers Discrete Tchebichef Transform (DTT) which offers better performance than the DCT due to its bitstream coding capabilities. This paper proposes a novel model of Discrete Tchebichef Transform (DTT) architecture with Register Pre-allocation based Folded Architecture (RPFA) for image compression. Through the cross-connection of folded architecture, the number of register usage is reduced. A Partial Cross Split Vedic Multiplier (PCSVM) method is introduced in the proposed DTT architecture. This multiplier design involves the cross function of the Vedic multiplier with the split pattern of multiplication binary stream. The optimal design of DTT architecture yields a minimum amount of FlipFlop (FF) counts, a latency and power consumption. The proposed PCSVM achieves higher Peak Signal to Noise Ratio (PSNR), better Structural Similarity Index (SSIM), lower delay, area, power consumption, Power-Delay Product (PDP), Mean Square Error (MSE) than the existing multiplier architectures. The proposed RPF-DTT architecture achieves a significant reduction in the resource consumption than the exact and approximate DTT architectures.

Hyperspectral image compression based on online learning spectral features dictionary

This paper proposes a novel method of lossy hyperspectral image compression using online learning dictionary. Spectral dictionary that learned in sparse coding mode could be used to represent the corresponding material. From the perspective of sparse coding, learning a sparse dictionary could achieve a better result of data decorrelation. In order to compress the hyperspectral data, an online learning sparse coding dictionary which could describe the characteristics of spectral curve was created to represent and reconstruct hyperspectral data. In the online learning phase, effective clustering algorithm is applied to generate and update the dictionary more properly. Results indicate that dictionary achieved by our method could improve the compression quality of hyperspectral image observably.

Low-Complexity Image and Video Coding Based on an Approximate Discrete Tchebichef Transform

The usage of linear transformations has great relevance for data decorrelation applications, like image and video compression. In that sense, the discrete Tchebichef transform (DTT) possesses useful coding and decorrelation properties. The DTT transform kernel does not depend on the input data and fast algorithms can be developed to real time applications. However, the DTT fast algorithm presented in literature possess high computational complexity. In this work, we introduce a new low-complexity approximation for the DTT. The fast algorithm of the proposed transform is multiplication-free and requires a reduced number of additions and bit-shifting operations. Image and video compression simulations in popular standards shows good performance of the proposed transform. Regarding hardware resource consumption for FPGA shows 43.1% reduction of configurable logic blocks and ASIC place and route realization shows 57.7% reduction in the area-time figure when compared with the 2-D version of the exact DTT.

Implementation of the DWT in a GPU through a Register-based Strategy

The release of the CUDA Kepler architecture in March 2012 has provided Nvidia GPUs with a larger register memory space and instructions for the communication of registers among threads. This facilitates a new programming strategy that utilizes registers for data sharing and reusing in detriment of the shared memory. Such a programming strategy can significantly improve the performance of applications that reuse data heavily. This paper presents a register-based implementation of the Discrete Wavelet Transform (DWT), the prevailing data decorrelation technique in the field of image coding. Experimental results indicate that the proposed method is, at least, four times faster than the best GPU implementation of the DWT found in the literature. Furthermore, theoretical analysis coincide with experimental tests in proving that the execution times achieved by the proposed implementation are close to the GPU’s performance limits.