Observation Window Size Research Articles

A novel probabilistic framework with interpretability for generator coherency identification

The increase of penetration of renewable energies has posed inevitable challenges to the stability and safety of power system operations, especially in large-scale multi-machine power systems. Emergency control is thereby crucial to avoid catastrophic accidents, and identifying coherent generators is the basis of wide-area control of a multi-machine power system. However, existing approaches are rule-based or rely on shallow machine learning, lacking effectiveness and robustness due to their insufficient ability of pattern mining from system monitoring indicators. To fill the gap, this paper proposes a novel end-to-end generator coherency identification framework, leveraging an improved auto-encoder to comprehensively exploit information of phasor measurement units (PMUs) obtained from wide-area measuring systems (WAMS). The framework jointly trains the feature extraction module and the clustering module to fully explore the shared knowledge and obtain cluster-specific representations. In addition, a visualization component is equipped with the process-agnostic framework for interpretability. Simulated and practical case studies validate the effectiveness of the proposed approach as it outperforms both deep learning baselines and state-of-the-art methods on all datasets under various situations, including observation window size changes, noisy data, or data missing at random.

Non‐contact monitoring of human heartbeat signals using mm‐wave frequency‐modulated continuous‐wave radar under low signal‐to‐noise ratio conditions

Radar has been shown potentially to be used for non-contact sensing of biosignals in a more comfortable and easier way than wearable and contact devices. By detecting and extracting body motions linked to physiological activities, accurate estimation of the heart rate is possible. However, the detection and estimation of heartbeat signal in practical applications is disturbed seriously by noise and respiration harmonic signal. In this study, a multi-level coarse-to-fine method is proposed for heartbeat rate (HR) estimation with a frequency-modulated continuous-wave radar operating at 60 GHz utilised to extract the heartbeat information of human body under low signal-to-noise ratio conditions. First, the frequencies, which are likely to be the HR, are extracted in the coarse estimation part. Then, evaluations about the extracted frequencies are conducted in the fine estimation part. Finally, the multiple signal classification algorithm is utilised to separate the HR frequency from the respiration harmonics, which solves the frequency resolution issue caused by the observation window size, and achieves high accuracy estimation of HR. The experiments on measured data demonstrate the effectiveness of the proposed method. Compared with the state of the art, the proposed method has superiority in terms of estimation accuracy of HR.

A Hierarchical Approach to Activity Recognition and Fall Detection Using Wavelets and Adaptive Pooling.

Human activity recognition has been a key study topic in the development of cyber physical systems and assisted living applications. In particular, inertial sensor based systems have become increasingly popular because they do not restrict users’ movement and are also relatively simple to implement compared to other approaches. In this paper, we present a hierarchical classification framework based on wavelets and adaptive pooling for activity recognition and fall detection predicting fall direction and severity. To accomplish this, windowed segments were extracted from each recording of inertial measurements from the SisFall dataset. A combination of wavelet based feature extraction and adaptive pooling was used before a classification framework was applied to determine the output class. Furthermore, tests were performed to determine the best observation window size and the sensor modality to use. Based on the experiments the best window size was found to be 3 s and the best sensor modality was found to be a combination of accelerometer and gyroscope measurements. These were used to perform activity recognition and fall detection with a resulting weighted F1 score of 94.67%. This framework is novel in terms of the approach to the human activity recognition and fall detection problem as it provides a scheme that is computationally less intensive while providing promising results and therefore can contribute to edge deployment of such systems.

Window-Based Morphometric Indices as Predictive Variables for Landslide Susceptibility Models

The identification of areas that are prone to landslides is essential in mitigating associated risks. This is usually achieved using landslide susceptibility models, which estimate landslide likelihood given local terrain conditions and the location of known past events. Detailed databases covering different conditioning factors are paramount in producing reliable susceptibility maps. However, thematic data from developing countries are scarce. As a result, susceptibility models often rely on morphometric parameters that are derived from widely-available digital elevation models. In most cases, simple parameters, such as slope, aspect, and curvature, computed using a moving window of 3 × 3 pixels, are used. Recently, the use of window-based morphometric indices as an additional input has increased. These rely on a user-defined observation window size. In this contribution, we examine the influence of observation window size when using window-based morphometric indices as core predictive variables for landslide susceptibility assessment. We computed a variety of models that include morphometric indices that are calculated with different window sizes, and compared the predictive capabilities and reliability of the resulting predictions. All of the models are based on the random forest algorithm. The results improved significantly when each window-based morphometric index was calculated with a different and meaningful observation window (AUC-ROC of 0.89 and AUC-PR of 0.87). The sensitivity analysis highlights both the highly-informative observation windows and the impact of their selection on the model performance. We also stress the importance of evaluating landslide susceptibility results while using different adapted metrics for predictive performance and reliability.

Botnet Identification in DDoS Attacks With Multiple Emulation Dictionaries

In a Distributed Denial of Service (DDoS) attack, a network (botnet) of dispersed agents (bots) sends requests to a website to saturate its resources. Since the requests are sent by automata, the typical way to detect them is to look for some repetition pattern or commonalities between requests of the same user or from different users. For this reason, recent DDoS variants exploit communication layers that offer broader possibility in terms of admissible request patterns, such as, e.g., the application layer. In this case, the malicious agents can pick legitimate messages from an emulation dictionary, and each individual agent sends a relatively low number of admissible requests, so as to make its activity non suspicious. This problem has been recently addressed under the assumption that all the members of the botnet use the same emulation dictionary. This situation is an idealization of what occurs in practice, since different clusters of agents are typically sharing only part of a global emulation dictionary. The diversity among the emulation dictionaries across different clusters introduces significant complexity in the botnet identification challenge. This work tackles this issue and provides the following main contributions. We obtain an analytical characterization of the message innovation rate of the DDoS attack with multiple emulation dictionaries. Exploiting this result, we design a botnet identification algorithm equipped with a cluster expurgation rule, which, under appropriate technical conditions, is shown to provide exact classification of bots and normal users as the observation window size increases. Then, an experimental campaign over real network traces is conducted to assess the validity of the theoretical analysis, as well as to examine the effect of a number of non-ideal effects that are unavoidably observed in practical scenarios.

A Novel and Reliable Framework of Patient Deterioration Prediction in Intensive Care Unit Based on Long Short-Term Memory-Recurrent Neural Network

The clinical investigation explored that early recognition and intervention are crucial for preventing clinical deterioration in patients in Intensive Care units (ICUs). Deterioration of patients is predictable and can be preventable if early risk factors are recognized and developed in the clinical setting. Timely detection of deterioration in ICU patients may also lead to better health management. In this paper, a new model was proposed based on Long Short-Term Memory-Recurrent Neural Network (LSTM-RNN) to predict deterioration of ICU patients. An optimisation model based on a modified genetic algorithm (GA) has also been proposed in this study to optimize the observation window, prediction window, and the number of neurons in hidden layers to increase accuracy, AUROC, and minimize test loss. The experimental results demonstrate that the prediction model proposed in this study acquired a significantly better classification performance compared with many other studies that used deep learning models in their works. Our proposed model was evaluated for two tasks: mortality and sudden transfer of patients to ICU. Our results show that the proposed model could predict deterioration before one hour of onset and outperforms other models. In this study, the proposed predictive model is implemented using the state-of-the-art graphical processing unit (GPU) virtual machine provided by Google Colaboratory. Moreover, the study uses a novel time-series approach, which is minute-by-minute. This novel approach enables the proposed model to obtain highly accurate results (i.e., an AUROC of 0.933 and an accuracy of 0.921). This study utilizes the individual and combined effectiveness of different types of variables (i.e., vital signs, laboratory measurements, GCS, and demographic data). In this study, data was extracted from MIMIC-III database. The ad-hoc frameworks proposed by previous studies can be improved by the novel and reliable prediction framework proposed in this research, which will result in predictions of more accurate performance. The proposed predictive model could reduce the required observation window (i.e., a reduction of 83%) for the prediction task while improving the performance. In fact, the proposed significant small size of observation window could obtain higher results which outperformed all previous works that utilize different sizes of observation window (i.e., 48 hours and 24 hours). Moreover, this research demonstrates the ability of the proposed predictive model to achieve accurate results (>80%) on ‘raw’ data in an experimental work. This shows that the rule-based pre-processing of clinical features is unnecessary for deep learning predictive models.

Development of intelligent control module for the J-TEXT electron cyclotron emission imaging system

An Electron Cyclotron Emission Imaging (ECEI) system has been developed on the J-TEXT tokamak and employed for the multi-dimensional electron temperature fluctuation imaging measurement in 2019 campaign. The intelligent control module has been developed and been used to flexibly setup the 256-channel ECEI system configurations for different plasma scenarios. The adjustments of key parameters are merged in this module including size and position of the geometric observation window, and the dynamic range of the signal and image resolution. The user-friendly interface is also developed for the tokamak operator to set the ECEI system focusing on physics interested region. This module is installed close to the diagnostic system around the J-TEXT tokamak, which can be remotely controlled by authorized operators in the control room. It significantly improves ECEI flexibility and stability, and also keeps the operation technician away from the harsh radiation environment.

Block Term Decomposition of ECG Recordings for Atrial Fibrillation Analysis: Temporal and Inter-Patient Variability

Responsible for 25% of strokes and 1/3 of hospitalizations due to cardiac related disturbances, atrial fibrillation (AF) is the most common sustained cardiac arrhythmia in clinical practice, considered as the last great frontier of cardiac electrophysiology. Its mechanisms are not completely understood, and a precise analysis of the atrial activity (AA) signal in electrocardiogram (ECG) recordings is necessary to better understand this challenging cardiac condition. Recently, the block term decomposition (BTD) has been proposed as a powerful tool to noninvasively extract AA in AF ECG signals. However, this tensor factorization technique was performed only in short ECG recordings and illustrated in single patients. To assess its performance and variability through different subjects, BTD is applied to a population of 10 AF patients in this paper. Also, its time variability is evaluated by means of long segments of AF ECG with varying observation window size. Experimental results show the consistency of BTD as an AA extraction tool, outperforming two well-known matrix-based methods in most of the observed cases for long and short AF ECG recordings.

A large-scale study of cultural differences using urban data about eating and drinking preferences

Traditional ways to study urban social behavior, e.g. surveys, are costly and do not scale. Recently, some studies have been showing new ways of obtaining data through location-based social networks (LBSNs), such as Foursquare, which could revolutionize the study of urban social behavior. We use Foursquare check-ins to represent user preferences regarding eating and drinking habits. Considering datasets differing in terms of volume of data and observation window size, our results indicate that spatio-temporal eating and drinking habits of users voluntarily expressed in LBSNs has the potential to explain cultural habits of the users. From this, we propose a methodology to identify cultural boundaries and similarities across populations at different scales, e.g., countries, cities, or neighborhoods. This methodology is extensively evaluated in several aspects. For instance, by proposing some variations of it disregarding some of the considered dimensions, as well as analyzing the results using datasets from different periods and window of observation. The results indicate that our proposed methodology is a promising approach for automatic cultural habits separation, which could enable new urban services.

Accounting for the Spatial Observation Window in the 2-D Fourier Transform Analysis of Shear Wave Attenuation

Recent measurements of shear wave propagation in viscoelastic materials have been analyzed by constructing the 2-D Fourier transform (2DFT) of the shear wave signal and measuring the phase velocity c(ω) and attenuation α(ω) from the peak location and full width at half-maximum (FWHM) of the 2DFT signal at discrete frequencies. However, when the shear wave is observed over a finite spatial range, the 2DFT signal is a convolution of the true signal and the observation window, and measurements using the FWHM can yield biased results. In this study, we describe a method to account for the size of the spatial observation window using a model of the 2DFT signal and a non-linear, least-squares fitting procedure to determine c(ω) and α(ω). Results from the analysis of finite-element simulation data agree with c(ω) and α(ω) calculated from the material parameters used in the simulation. Results obtained in a viscoelastic phantom indicate that the measured attenuation is independent of the observation window and agree with measurements of c(ω) and α(ω) obtained using the previously described progressive phase and exponential decay analysis.

Burst topic discovery and trend tracing based on Storm

With the rapid development of the Internet and the promotion of mobile Internet, microblogs have become a major source and route of transmission for public opinion, including burst topics that are caused by emergencies. To facilitate real time mining of a large range of burst topics, in this paper, we proposed a method to discover burst topics in real time and trace their trends based on the variation trends of word frequencies. First, for the variation trend of the words in microblogs, we adopt a non-homogeneous Poisson process model to fit the data. To represent the heat and trend of the words, we introduce heat degree factor and trend degree factor and realise the real time discovery and trend tracing of the burst topics based on these two factors. Second, to improve the computing performance, this paper was based on the Storm stream computing framework for real time computing. Finally, the experimental results indicate that by adjusting the observation window size and trend degree threshold, topics with different cycles and different burst strengths can be discovered.

BER Analysis of Decision-Feedback Multiple-Symbol Detection in Noncoherent MIMO Ultrawideband Systems

In this paper, we investigate noncoherent multiple-input-multiple-output (MIMO) ultrawideband (UWB) systems where the signal is encoded by differential space-time block code (DSTBC). DSTBC enables noncoherent MIMO UWB systems to achieve diversity gain. However, the traditional noncoherent symbol-by-symbol differential detection (DD) for DSTBC-UWB suffers from performance degradation compared with coherent detection. We introduce a noncoherent multiple-symbol detection (MSD) scheme to enhance the performance of DSTBC-UWB systems. Although the MSD scheme can boost the performance more as the observation window size becomes larger, the complexity of the exhaustive search for MSD also exponentially increases in terms of the window size. To decrease the computational complexity, the concept of decision feedback (DF) is introduced to the MSD for DSTBC-UWB in this paper. The resultant DF-MSD yields reasonable complexity and solid performance improvement. We provide the bit-error-rate analysis for the proposed DF-MSD. Both theoretical analysis and simulation results validate the proposed scheme.

Band-limited angular spectrum numerical propagation method with selective scaling of observation window size and sample number

Band-limited angular spectrum (BLAS) methods can be used for simulating the diffractional propagation in the near field, the far field, the tilted system, and the nonparaxial system. However, it does not allow free sample interval on the output calculation window. In this paper, an improved BLAS method is proposed. This new algorithm permits a selective scaling of observation window size and sample number on the observation plane. The method is based on the linear convolution, which can be calculated by fast Fourier transform effectively.

Cramér–Rao lower bound for non-data-aided carrier phase estimation from general M-ary phase-shift keying modulation signals over flat Rayleigh fading channel

The authors derive the true Cramér–Rao lower bound (CRLB) for a non-data-aided (NDA) carrier phase estimation from general M-ary phase-shift keying (M-PSK) modulation signal over flat Rayleigh fading channels (FRFC). The numerical Monte Carlo method is used to compute the true CRLB from general M-PSK modulation signals over FRFC with complex additive white Gaussian noise. The efficiency of the proposed method is made obvious through the assessment of the true CRLBs from both symmetric and asymmetric 8-PSK modulation signals for various observation window sizes. The simulation results show that optimal NDA carrier phase recovery is harder to acquire from symmetric modulation signals for signal-to-noise ratio (SNR) range up to 5 dB, but depends on the constellation density, only for SNR range superior to 5 dB. For both constellations the CRLB decreases as long as the observation window size increases. The derived bound provides an efficient standard for evaluating the performance of any unbiased NDA carrier phase estimator from general M-PSK modulation signals over FRFC. The method presented here stands useful for general M-ary quadrature amplitude modulation signals as well.

Decision-feedback multiple symbol detection for differential space-time block coded UWB systems

The symbol-by-symbol differential detection scheme, originally employed to detect differential space-time block coded (DSTBC) ultra-wideband (UWB) signal in previous work, suffers from a performance degradation compared with coherent detection. In this paper, a noncoherent multiple symbol detection (MSD) scheme is proposed to enhance the performance of DSTBC-UWB system. After the signal model of DSTBC-UWB is analyzed, the optimal maximum likelihood (ML) MSD is derived and then a suboptimal ML-MSD is obtained by channel approximating. However, exhaustive search based MSD is impractical in DSTBC-UWB systems since its complexity increases exponentially with observation window size. To decrease the computational complexity, the decision-feedback (DF) scheme is introduced to MSD. The resultant DF-MSD yields reasonable complexity and solid performance improvement.

LONG MEMORY IN STOCK MARKET VOLATILITY: THE INTERNATIONAL EVIDENCE

It is still a hot topic to catch the auto-dependence behavior of volatility. Here, based on the measurement of average volatility, under different observation window size, we investigated the dependence of successive volatility of several main stock indices and their simulated GARCH(1, 1) model, there were obvious linear auto-dependence in the logarithm of volatility under a small observation window size and nonlinear auto-dependence under a big observation. After calculating the correlation and mutual information of the logarithm of volatility for Dow Jones Industrial Average during different periods, we find that some influential events can change the correlation structure and the volatilities of different periods have distinct influence on that of the remote future. Besides, GARCH model could produce similar behavior of dependence as real data and long memory property. But our analyses show that the auto-dependence of volatility in GARCH is different from that in real data, and the long memory is undervalued by GARCH.

Multiple-Symbol differential detection for distributed space-time block coding cooperative system

In distributed space-time coding cooperative systems, differential detection gains much attention because it needs no channel state information at the receivers. But conventional differential detection performs about 3dB loss than coherent detection. In this paper, we propose a multiple-symbol differential detection algorithm for relay networks based on maximum-likelihood principle. The simulation results show that the multiple-symbol differential detection can narrow the performance gap between the conventional differential detection and coherent detection for Alamouti and square real orthogonal coding systems with corresponding numbers of relays, especially in high signal-tonoise ratio regime. And with the increase of observation window size, the Doppler effect on performance is decreased and the error floor encountered in conventional differential detection can be eliminated provided that the value of observation window size is sufficiently high.

The Comparison of Machine Learning Algorithms on Online Classification of Network Flows

Online classification of network flows is a process that captures packets generated by network applications and identifies types of network applications (or flows) in real time. There are three key issues about online classification: observation window size, feature selection, and classification algorithms. In this paper, by collecting five types of typical network flow data as the experiment sample data, the authors found observation window size 7 is the best for the sample data and most classifiers. The authors proposed a full feature set based on the standard feature set which reflects statistical features of network flows. Using five commonly used feature selection methods, the authors identified the most effective features could be reduced from 56 original features to 11 effective features. Lastly, according to special need for online classification, the authors studied 11 different classifiers on their classification accuracy, model construction time, and classification speed. The results show that C4.5 and JRip are the two best algorithms for online classification.

Sphere Decoding Based Multiple Symbol Detection for Differential Space-Time Block Coded Ultra-Wideband Systems

In this letter, a sphere decoding (SD) based noncoherent multiple symbol detection (MSD) scheme is proposed for differential space-time block coded (DSTBC) ultra-wideband (UWB) impulse radio systems. Exhaustive search based maximum likelihood (ML) MSD is impractical in DSTBC-UWB systems since its complexity exponentially increases with observation window size. To decrease the computational complexity, the original MSD metric is transformed into another equivalent form which can be implemented by SD for DSTBC-UWB systems.

Fast multiple-symbol detection for free-space optical communications

In this paper, we investigate noncoherent detection, i.e. detection assuming the absence of channel state information at the receiver, of on-off keying in an intensity modulation and direct detection (IM/DD) free-space optical (FSO) system. To partially recover the performance loss associated with symbol- by-symbol noncoherent detection, we consider the application of multiple-symbol detection (MSD), in which block-wise decisions are made using an observation window of several bit intervals. Specifically, we develop a fast search algorithm for optimal MSD and propose a reduced-complexity decision metric suitable for suboptimal MSD; performance results confirm that the optimal and suboptimal metrics perform comparably well. Significantly, the complexity of our receiver, on a per bit-decision basis, is only logarithmically dependent on the observation window size. We also present the framework for a decision-feedback receiver and obtain performance expressions for the ideal case of error-free feedback; these expressions serve as an upper bound to the performance of MSD. Analytical and simulation results indicate that as the observation window size increases, the performance of the MSD receiver approaches that of detection with channel state information. The conclusion is reached that the proposed implementation provides an attractive low-complexity mechanism for performing noncoherent MSD in FSO systems.