Overlap Aware Compressed Signal Classification

Compressive sensing (CS) eases signal acquisition by relaxing the barrier of Nyquist rate. It randomly samples the signal at a rate proportional to the signal sparsity. This compresses the signal at the time of sensing. This superiority of CS is hindered by the complex reconstruction algorithms. The way-out is proposed by an advanced area of CS, in which signal processing is directly done on compressive measurements, avoiding reconstruction. This area is known as compressed signal processing (CSP). It saves power and time and also reduces the computational complexity drastically. CSP is best suited to real-time condition monitoring scenarios where devices are power constrained. These devices generate massive amount of unlabeled data, labeling of which is done manually to train a classifier for condition monitoring. This is quite laborious and time-consuming task. To overcome this burden, use of informed instance selection scheme of active learning (AL) is proposed in this paper. Using this technique, the classifier can be trained even if the data have only few labeled instances. Thus, combining CSP with AL improves the performance of condition monitoring systems to a great extent. The simulation results show that the significant improvement can be obtained using the proposed method.

Most of the recent compressive sensing (CS) literature has focused on sparse signal recovery based on compressive measurements. However, exact signal recovery may not be required in certain signal processing applications such as in inference problems. In this paper, we provide performance limits of classification of sparse as well as not necessarily sparse signals based on compressive measurements. When signals are not necessarily sparse, we show that Kullback-Leibler and Chernoff distances between two probability density functions under any two hypotheses are preserved up to a factor of M/N with M(<;N)-length compressive measurements compared to that with N-length original measurements when the pdfs of the original-length observation vectors exhibit certain properties. These results are used to quantify the performance limits in terms of upper and lower bounds on the probability of error in signal classification with M-length compressive measurements. When the signals of interest are sparse in the standard canonical basis, performance limits are derived in terms of lower bounds on the probability of error in classifying sparse signals with any classification rule.

In this paper, we propose a novel direction-of-arrival (DOA) estimation scheme, which is named the compressive one-bit measurement scheme. In the proposed scheme, the one-bit quantization technique is used to reduce the system cost in terms of the analog-to-digital converter (ADC). However, the one-bit quantization leads to a serious information loss, thus compromising the estimation accuracy. Inspired by the compressive sensing theory, the compressive measurement method is used to expand the receive array aperture. To be specific, the compressive measurement allows more sensors than the number of front-end circuit chains to be used. Thus, the additional information is obtained, and the estimation performance can be improved from the system structure layer. It is noted that by introducing the compression operation, the relationship between the normalized and the original covariance matrix is different to the conventional one-bit quantization structure. Thus, two DOA estimation algorithms, i.e., the iterative compressive measurement-based multiple signal classification (CM-MUSIC) and the compressive sensing (CS)-based algorithm, are proposed for the compressive one-bit measurement scheme. The iterative CM-MUSIC can achieve a high-resolution estimation, whereas the CS-based algorithm can resolve more sources than the sensors. Numerical simulations validate the effectiveness of the proposed scheme.

Wideband spectrum sensing is a challenging task due to the constraints of digital signal processing (DSP) unit using in extant wireless systems. Compressive sensing (CS) is a new paradigm in signal processing, chosen for sparse wideband spectrum estimation with compressive measurements, thus provides relief of high-speed DSP requirements of cognitive radio (CR) receivers. In CS, whole wideband spectrum is estimated to find an opportunity for a CR usage requiring significant computation as well as sensing time, hence shrinkage the achievable throughput of CRs. In this paper, a novel model based CR receiver wideband sensing unit is addressed where a significant portion of the wideband spectrum is approximated through compressive sensing rather than recovering the total wideband spectrum. This model necessitates lesser sensing time and lower computational burden to detect a signal and as a result a level up of throughput is obtained. As a result, the sensing time gain improves the achievable throughput of the CRs which reflects on the simulation results and testifies the effectiveness of the proposed model. Therefore, a reduction of computational complexity is addressed without interfering with the detection performances, evaluated after spectrum estimation of a preferred band of interest by means of a well-known energy detector.

This paper presents modification of the TwIST algorithm for Compressive Sensing MRI images reconstruction. Compressive Sensing is new approach in signal processing whose basic idea is recovering signal form small set of available samples. The application of the Compressive Sensing in biomedical imaging has found great importance. It allows significant lowering of the acquisition time, and therefore, save the patient from the negative impact of the MR apparatus. TwIST is commonly used algorithm for 2D signals reconstruction using Compressive Sensing principle. It is based on the Total Variation minimization. Standard version of the TwIST uses masked 2D Discrete Fourier Transform coefficients as Compressive Sensing measurements. In this paper, different masks and different transformation domains for coefficients selection are tested. Certain percent of the measurements is used from the mask, as well as small number of coefficients outside the mask. Comparative analysis using 2D DFT and 2D DCT coefficients, with different mask shapes is performed. The theory is proved with experimental results.

We investigate the geometric complexity of decision boundaries for robust training compared to standard training. By considering the local geometry of nearest neighbour sets, we study them in a model-agnostic way and theoretically derive a lower-bound \(R^*\in \mathbb {R}\) on the perturbation magnitude \(\delta \in \mathbb {R}\) for which robust training provably requires a geometrically more complex decision boundary than accurate training. We show that state-of-the-art robust models learn more complex decision boundaries than their non-robust counterparts, confirming previous hypotheses. Then, we compute \(R^*\) for common image benchmarks and find that it also empirically serves as an upper bound over which label noise is introduced. We demonstrate for deep neural network classifiers that perturbation magnitudes \(\delta \ge R^*\) lead to reduced robustness and generalization performance. Therefore, \(R^*\) bounds the maximum feasible perturbation magnitude for norm-bounded robust training and data augmentation. Finally, we show that \(R^*< 0.5R\) for common benchmarks, where R is a distribution’s minimum nearest neighbour distance. Thus, we improve previous work on determining a distribution’s maximum robust radius.

Compressive sensing (CS) is a viable source of an innovative 5G system, what's more, it's an effective technology to deal with the data redundancy problem of massive machine-to-machine communication (MMC), since it enables the recovery of sparse and approximately sparse signals with significantly fewer samples than demanded by Nyquist-Shannon sampling theory. Interference in signal will lead a series problem to signal processing, which will be a recovery error proportional to the interference energy in CS. This paper tries to mitigate interference by proposing a compressive interference pre-filter based on linear minimal mean square error (LMMSE). The main contributions are that two practical estimation methods that based on autocorrelation function and auto-covariance function respectively, are applied to estimate the LMMSE-based filtering matrix. The estimation and pre-filter algorithms can be practically integrated into the compressive signal processing framework with improved robust property to noise and interference.

The inverter is the most vulnerable module of photovoltaic (PV) systems. The insulated gate bipolar transistor (IGBT) is the core part of inverters and the root source of PV inverter failures. How to effectively diagnose the IGBT faults is critical for reliability, high efficiency, and safety of PV systems. Recently, deep learning (DL) methods are widely used for fault detection and diagnosis. Different from traditional diagnosis methods, DL methods use deep neural networks which can automatically extract the useful representative features from raw data. However, DL methods require large amounts of data, which leads to the high cost of communication, storage, and computation. To tackle these issues, a data-driven fault detection and diagnosis method for IGBT open-circuit faults based on compressed sensing (CS) and convolutional neural networks (CNN) is proposed in this paper. CS is adopted to compress raw signals, and the optimal value of compression ratio (CR) is determined by considering the trade-off between classification accuracy and model training time. The overlap sampling method is adopted for data segmentation. Meanwhile, overlap sampling can also increase the number of training samples and improve the sample correlation. The compressed signals are segmented and reconstructed into two-dimensional feature maps for model training. Finally, compared with CNN of the same structure, the developed CS-CNN model can compress 85% of data without accuracy loss. The performance comparison with the state-of-the-art networks demonstrates that the test accuracy is 98.68% and the model training time is much shorter than other methods.

Capturing images in thousands of contiguous spectral bands has been made simpler with the emergence of technology in the field of hyperspectral remote sensing. Despite of these huge data available for analysis, Hyperspectral images (HSI) face many challenges due to high dimensionality, noise, spectral mixing and computational complexity. Several preprocessing methods can be used to overcome the above mentioned issues. In this paper, an enhancement technique using 2D-Empirical Wavelet Transform (EWT) is used as a preprocessing step for the HSI reconstruction prior to sparsity based classification (Subspace Pursuit and Orthogonal Matching Pursuit). The effectiveness of the proposed method is proved by comparing the classification results obtained with and without applying preprocessing. Experimental analysis shows a significant improvement in the classification accuracies i.e., for 40 $$\%$$ of training samples, OMP shows an improvement in overall classification accuracy from 66.12 $$\%$$ to 93.20 $$\%$$ and SP shows an improvement from 66.36 $$\%$$ to 92.74 $$\%$$ .

Hyperspectral data compression has received considerable interest in recent years. Contrarily to the conventional compression schemes, which first acquire the full data set and then implement some compressing technique, compressive sensing (CS) acquires directly the compressed signal which will be later recovered on the ground station. The CS paradigm fits perfectly the requirements of onborad hyperspectral imaging systems in terms of energy, computing power, and bandwidth. By using CS in these systems, the amount of data acquired and transmitted to the ground stations is reduced and the bulk of the computation to infer the original data is carried out in the ground stations. In this paper, we present a new technique to perform CS of hyperspectral images (HSIs), which exploit the fact that HSIs admits a low dimensional linear representation. The proposed method is blind in the sense that linear representation is learned with low computational cost from the compressed measurements. Furthermore the proposed method is very light from the computational point of view and it can recover perfectly the original image in noise-free scenarios. The effectiveness of the proposed method is illustrated in both synthetic and real scenarios.

Recently, there are a lot of study of direction of arrival (DoA) estimation using compressive sensing (CS). As CS is a new paradigm in signal processing, there are many aspects of this method that can be investigated. In the case of DoA estimation in noisy measurement, it is important to correctly determine a correct threshold of CS reconstruction, particularly when CS reconstruction is implemented using L 1 -norm minimization. Too small threshold value will make the correct DoA does not lies in CS reconstruction searching area, while too large threshold value will burden CS iteration to select a solution from a large number of possible solutions. In this paper, we derived an optimal threshold value for CS reconstruction for DoA estimation mathematically and verified the result using computer simulation. Using Gaussian noise model, we obtain the chi-square distribution of euclidean distance of noisy and noiseless received vector. We introduce the thresholding index κ to scale the standard deviation of chi-square distribution to determine the CS reconstruction threshold and simulate this value for various SNR. We find that the optimal κ value 0.5 to 1 for high noise environment, and optimal κ value 1 to 2 in low noise environment.

Compressed Sensing (CS) theory is a new method of signal acquisition and processing proposed in recent years. With small amount of sampling data recovering original data to precisely reconstruct sparse signal or compression signal, the theory breaks though the restriction of Nyquist sampling theorem. CS can avoid enormous sampling data waste but also reduce the complexity of image coding. This paper reviews the basic theory of CS and its three key points, including signal sparse representation, design of measurement matrix and reconstruction algorithms. Then, the application of CS in the field of remote sensing image compression technology is studied. Using MATLAB software, we do a series of CS emulation experiments compared with the traditional compression methods. The results show that the proposed method has a good performance on the remote sensing image compression.

Electrocorticography (ECoG) is a type of electrophysiological monitoring useful for recording the activity from the cerebral cortex. It has emerged as a promising recording technique in brain-computer interfaces (BCI). Compression of these signals is essential for saving power and bandwidth in the novel application scenarios of Health-based IoT and Body Area Networks. However, this task is particularly challenging since, ECoG signals are not compressible either in time domain or in frequency domain. To that end, Block Sparse Bayesian Learning (BSBL) techniques were suggested for the reconstruction of compressed EEG and ECG signals, which is however, computationally demanding. Furthermore, given the heterogeneity in modern computing systems, careful design partitioning is required to most effectively evaluate the particular resources available on the deployed architecture. In this paper, we propose to utilise a combination of compressive sensing and neural network for the compression and reconstruction of ECoG signals, respectively. For the choice of the neural network, a multi-layer perceptron regressor with a stochastic gradient descent solver is developed. For a sample system, we show that the network has a compression ratio of 50%, and reconstruction accuracy of 89.85% after training with a practical, medium-sized dataset. In general, the results show that the most efficient system implementation is a heterogeneous architecture combining a CPU and a fieldprogrammable gate array (FPGA).

Compressed Sensing (CS) is a promising method that recovers the sparse and compressible signals from severely under-sampled measurements. CS can be applied to wireless communication to enhance its capabilities. As this technology is proliferating, it is possible to explore its need and benefits for emerging applicationsCompressive Sensing for Wireless Communication provides: • A clear insight into the basics of compressed sensing • A thorough exploration of applying CS to audio, image and computer vision • Different dimensions of applying CS in Cognitive radio networks • CS in wireless sensor network for spatial compression and projection • Real world problems/projects that can be implemented and tested • Efficient methods to sample and reconstruct the images in resource constrained WMSN environmentThis book provides the details of CS and its associated applications in a thorough manner. It lays a direction for students and new engineers and prepares them for developing new tasks within the field of CS. It is an indispensable companion for practicing engineers who wish to learn about the emerging areas of interest.

Recently, compressive sensing (CS) schemes have been studied as a new compression modality that exploits the sensing matrix in the measurement scheme and the reconstruction scheme to recover the compressed signal. In addition, CS is exploited in medical imaging (MI) to support efficient sampling, compression, transmission, and storage of a large amount of MI. Although CS of MI has been extensively investigated, the effect of color space in CS of MI has not yet been studied in the literature. To fulfill these requirements, this article proposes a novel CS of MI based on hue-saturation value (HSV), using spread spectrum Fourier sampling (SSFS) and sparsity averaging with reweighted analysis (SARA). An HSV loop that performs SSFS is proposed to obtain a compressed signal. Next, HSV–SARA is proposed to reconstruct MI from the compressed signal. A set of color MIs is investigated, such as colonoscopy, magnetic resonance imaging of the brain and eye, and wireless capsule endoscopy images. Experiments were performed to show the superiority of HSV–SARA over benchmark methods in terms of signal-to-noise ratio (SNR), structural similarity (SSIM) index, and measurement rate (MR). The experiments showed that a color MI, with a resolution of pixels, could be compressed by the proposed CS at MR of 0.1, and could be improved in terms of SNR being 15.17% and SSIM being 2.53%. The proposed HSV–SARA can be a solution for color medical image compression and sampling to improve the image acquisition of medical devices.