Sequential Nonparametric Algorithm for Detecting Time Series Breakdown

Community detection is a dominant research field in social media mining. Numerous algorithms have been developed to detect community structure in network. Each of them focuses on different characteristic of data and generates different communities for same dataset. So cluster validation or comparative analysis of these algorithms is required to compare the efficiency of these algorithms. There are two known ways for cluster validation: Internal Quality Measure (IQM) and External Quality Measure (EQM). IQM evaluates the cluster without reference to external information while EQM requires already known clusters i.e. ground truth for comparison. Most of the researchers have compared Community Detection algorithms on the basis of either IQM or EQM but not both. In this paper, we have tried to compare nine existing sequential community detection algorithms on the basis of both IQM and EQM and we observed that overall performance of Louvain, Fastgreedy and Walktrap algorithm is more efficient than other algorithms. From this paper, researchers working in this area can get information about major sequential community detection algorithms and their efficiency on small real world datasets. They can incorporate the findings to develop more efficient algorithms.

Recommender systems are widely used in electronic commerce, social media and online streaming services to provide personalized recommendations to the users by exploiting past ratings and interactions. This paper considers the security aspect with quick and accurate detection of attacks by observing the newly created profiles sequentially to prevent the damage which may be incurred by the injection of new profiles with dishonest ratings. The proposed framework consists of a latent variable model, which is trained by a variational EM algorithm, followed by a sequential detection algorithm. The latent variable model generates homogeneous representations of the users given their rating history and mixed data-type attributes such as age and gender. The representations are then exploited to generate univariate statistics to be efficiently used in a CUSUM-like sequential detection algorithm that can quickly detect persistent attacks while maintaining low false alarm rates. We apply our proposed framework to three different real-world datasets and exhibit superior performance in comparison to the existing baseline algorithms for both attack profile and sequential detection. Furthermore, we demonstrate robustness to different attack strategies and configurations.

Slotted waveguide antennas are widely used in radio engineering systems for various purposes due to their advantages: high directivity, the possibility of constructing linear and lattice radiators, compactness and manufacturability. Despite the advantages, designing slotted waveguide antennas is a difficult task. The problem is that the antenna characteristics are determined not only by the electrodynamic processes occurring inside the waveguide, but also by the parameters of the slots, their location and orientation. In this case, it is necessary to take into account the mutual influence of the slots on each other, as well as the matching of the antenna with the supply path. In particular, resonant and non-resonant slotted waveguide antennas are distinguished. In resonant designs, the slots are excited by a standing wave, which ensures high efficiency, but imposes strict matching requirements. Traveling wave is used in non-resonant antennas, which makes it possible to expand the band and simplify matching, however, the task arises of distributing the characteristics along the structure. An important task in the design is the formation of a directional pattern with inclination of the main lobe and sidelobe level. These parameters are determined by the amplitude-phase distribution and depend on the location and characteristics of the slots. The correct choice of their size and position allows you to control the orientation and achieve the required characteristics. The article presents a scientific justification of the mathematical apparatus for calculating gaps for optimization and obtaining the characteristics. The algorithm takes into account the processes in the waveguide, the mutual influence and the alignment with the supply path. The technique involves numerical modeling, followed by analysis of the results and optimization of parameters to achieve the tilt of the main lobe, reduce the level of the side lobes and provide the electrical characteristics. Purpose – development and optimization of a slotted waveguide antenna with a given inclination of the main lobe of the radiation pattern and the required sidelobe level based on the slot calculation algorithm. A slotted waveguide antenna designed to operate at a frequency of 10 GHz is calculated based on an algorithm for determining slot conductivities at equal-amplitude current distribution. A non-resonant slotted waveguide antenna with variable-phase slots in a waveguide with an absorbing load is considered. Numerical modeling in the FEKO has been performed, confirming the correctness of the calculations. Optimization of the design parameters in DT Seven has been carried out to form a directional pattern with a given inclination of the main lobe and the required sidelobe level. Radiation characteristics meeting the set requirements have been obtained. A universal algorithm has been developed for calculating a non-resonant slotted waveguide antenna with variable-phase slots, which allows forming a radiation pattern with specified parameters. The possibility of optimization makes the method adaptable to different requirements and ensures wide applicability in practice. The implemented approach allows us to expand it to calculate antennas with nonuniform amplitude current distribution, as well as other types of slotted waveguide antennas.

A sequential two-dimensional (2D) partial response maximum likelihood (PRML) detection scheme for holographic data storage (HDS) systems is proposed. We use two complexity reduction schemes, a reduced-state trellis and a constant-weight (CW) constraint. In the reduced-state trellis, the limited candidate bits surrounding the target bit are considered for the 2D PRML detector. In the CW constraint, the trellis transitions that violate the CW condition that each code-word block has only one white bit are eliminated. However, the 2D PRML detector using the complexity reduction schemes, which operates on 47 states and 169 branches, has performance degradation. To overcome performance degradation, a sequential detection algorithm uses the estimated a priori probability. By the sequential procedure, we mitigate 2D intersymbol interference with an enhanced reliability of the branch metric. Simulation results show that the proposed 2D PRML detection scheme yields about 3 dB gains over the one-dimensional PRML detection scheme.

Motivated from a molecular dynamics context we propose a sequential change point detection algorithm for vector-valued autoregressive models based on Bayesian model selection. The algorithm does not rely on any sampling procedure or assumptions underlying the dynamics of the transitions and is designed to cope with high-dimensional data. We show the applicability of the algorithm on a time series obtained from numerical simulation of a penta-peptide molecule.

Classification of multivariate time series data, often including both time points and intervals at variable frequencies, is a challenging task. We introduce the KarmaLegoSification (KLS) framework for classification of multivariate time series analysis, which implements three phases: (1) application of a temporal abstraction process that transforms a series of raw time-stamped data points into a series of symbolic time intervals; (2) mining these symbolic time intervals to discover frequent time-interval-related patterns (TIRPs), using Allen’s temporal relations; and (3) using the TIRPs as features to induce a classifier. To efficiently detect multiple TIRPs (features) in a single entity to be classified, we introduce a new algorithm, SingleKarmaLego, which can be shown to be superior for that purpose over a Sequential TIRPs Detection algorithm. We evaluated the KLS framework on datasets in the domains of diabetes, intensive care, and infectious hepatitis, assessing the effects of the various settings of the KLS framework. Discretization using Symbolic Aggregate approXimation (SAX) led to better performance than using the equal-width discretization (EWD); knowledge-based cut-off definitions when available were superior to both. Using three abstract temporal relations was superior to using the seven core temporal relations. Using an epsilon value larger than zero tended to result in a slightly better accuracy when using the SAX discretization method, but resulted in a reduced accuracy when using EWD, and overall, does not seem beneficial. No feature selection method we tried proved useful. Regarding feature (TIRP) representation, mean duration performed better than horizontal support, which in turn performed better than the default Binary (existence) representation method.

We consider the problem of spectrum sensing in cognitive radio networks under model uncertainties, i.e., the distribution under null hypothesis and the alternate hypothesis is not known. We model the presence of the primary signal as a mean change problem. Two simple nonparametric sequential algorithms for testing the mean are proposed. These nonparametric algorithms are applied to a decentralized detection framework. These algorithms show better performance than some of the existing nonparametric tests in literature. The performance of these algorithms is analysed via theory and simulations.

We describe a fully automated method for quantification of left ventricular performance by equilibrium radionuclide ventriculographic studies, based on subdivision of the left ventricular region into 9 equiangular sectors. The precise identification of the left ventricular contours is achieved by the use of morphological and functional criteria in a sequential edge detection algorithm with a success rate of 96%. In addition to left ventricular global and sectorial ejection fraction the first harmonic of the corresponding Fourier spectrum is approximated to each sectorial time-activity curve and to the global one. Sectorial phase is calculated as the difference between the phase of the sectorial and global first Fourier component. Computerized comparison between the sectorial parameters at rest and during peak exercise localizes and classifies the degree of global and regional impairment in response to exercise. The processing time of 60 sec makes this method suitable for routine use. The validity of our procedure has been tested in 34 patients before and after successful transluminal coronary angioplasty. In these patients, 73% of the stenosed vessels before dilatation were localized by sectorial ejection fraction, 77% by sectorial phases, and 88% by the combination of both.

Cognitive Radio is an enabling technology that is expected to address the problem of scarcity of spectrum by intelligent access, mobility and management of spectrum and pave way for efficient usage of the same. Increasingly, it is seen that most research is directed towards building theoretic models for spectrum access and management. Though some of these are backed by simulations, certainty of any method can be proven only by practical implementation on hardware testbeds in real-world environments. Sometimes, assumptions that are made to satisfy simulation requirements are difficult to realize in practical scenarios, thereby, their implementation in real-world is even more challenging. In this paper, taking cue from our earlier implementation work on simple Snapshot ED and CUSUM based algorithms for single node spectrum sensing, we explain the implementation of Sequential change detection algorithm in co-operative mode, the Dual-CUSUM. Single node sensing is not reliable, especially, when the nodes are subjected to shadow fading due to obstacles (hidden node problem). This is mitigated by using co-operative sensing that exploits the inherent spatial diversity and it is seen that with certain fusion rules, Cooperative CUSUM (aka Dual-CUSUM) is better than its counterpart Co-operative Snapshot ED, especially, under low SNR regimes.

Electronic warfare (Polish acronym WRE ‒ walka radioelektroniczna) is a set of militaryactions based on the use of the electromagnetic spectrum and it includes:— reception and identification of electromagnetic emissions,— reducing the effectiveness of the enemy’s electronic devices,— enabling effective use of the spectrum by own forces.The task of the WRE in peacetime is to obtain information about radio-electronic devices, especiallyradars. These tasks are performed by ELINT (Electronic Intelligence) and/or ESM (Electronic WarfareSupport Measures) devices. They operate passively in the 0.5÷18 GHz band, their detection is difficultand the information they provide allows to detect a threat to own forces and to take appropriatecountermeasures. Radar recognition is a complex process that involves the following stages. The firststage: gathering measurements and their grouping into the so-called packets and graphical display ofthe received pulse trains. The measurements are represented by the so-called measurement vectors(descriptors) containing signal parameters, including the TOA (Time of Arrival) pulse with nanosecondaccuracy. The second stage: associating pulse packets with logical sources in order to obtainrepresentative data strings. This stage is signal sorting and deinterleaving. The third stage: estimationof the WS signal vector based on the associated pulse packets. The final stage of data processing is thecomparison of the obtained signal descriptor parameters WS with Database (BD) radar patterns. Thereconnaissance result allows for the assessment of the threat resulting from the operation of this radar.Radar recognition requires precise determination of many parameters of its signal. For pulsed radars,important parameters are the Pulse Repetition Interval (PRI) and the type of PRI modulation. Radarsare characterised by a high complexity of PRI changes. This provides great utility and makes it possibleto distinguish even radar units and changes in their locations. This allows for the detection and fullidentification of objects that use radars (e.g. ships) with the given object completely out of sight; it alsoallows to predict intentions of the target. The basic inter-pulse PRI modulations include: stagger, dwelland switch, jitter or sliding. PRI changes in radar signals may be periodic with long cycles, difficultto identify in cluttered conditions. Unfortunately, the received pulse trains are usually distorted. Thismakes the WS estimation process difficult ‒ especially in automatic mode. PRI analysis has receivedmuch attention in the literature. The methods are computationally complex and have numerouslimitations. This article presents ASWC algorithm (Sequential Cycle Detection Algorithm) of PRIproprietary, with relatively low computational complexity. Examples of test results confirming its higheffectiveness in automatic PRI analysis performed under interference conditions were also presented.Keywords: radars, electronic warfare, radar recognition, ELINT, ESM, pulse descriptor, PRI

In this paper, we propose a secure framework for healthcare monitoring using the Internet of Medical Things (IoMT). In spite of their deployment, these devices still vulnerable to several cyber-attacks, ranging from unauthorized access to private medical data to data modification and injection. These attacks can compromise the privacy of the monitored patient, reduce the reliability of the monitoring system and may harm the life of monitored patient. In this paper, we propose a new framework to detect attacks and secure the communications in IoMT. To prevent eavesdropping and modification attacks, we propose the Ephemeral Elliptic Curves Diffie-Hellman (EECDH) to derive a session key used to provide confidentiality and authenticity. To detect injected measurements, flooding triggered by compromised devices and medical changes in physiological data, we applied the sequential change point detection algorithm Pruned Exact Linear Time (PELT) followed by the boxplot. Our experimental results show that our approach is able to increase the reliability and the accuracy of remote monitoring system, while reducing the false alarms triggered by injected measurements.

The amount of data generated by online social networks such as Facebook, Twitter, etc., has recently experienced an enormous growth. Extracting useful information such as community structure, from such large networks is very important in many applications. Community is a collection of nodes, having dense internal connections and sparse external connections. Community detection algorithms aim to group nodes into different communities by extracting similarities and social relations between nodes. Although, many community detection algorithms in literature, they are not scalable enough to handle large volumes of data generated by many of the today's big data applications. So, researchers are focusing on developing parallel community detection algorithms, which can handle networks consisting of millions of edges and vertices. In this article, we present a comprehensive survey of parallel community detection algorithms, which is the first ever survey in this domain, although, multiple papers exist in literature related to sequential community detection algorithms.

In this paper, we propose a sequential spectrum sensing algorithm for cognitive radio systems, which we term the sequential shifted chi-square test (SSCT). SSCT has the following attractive features for practical implementations. First, SSCT employs a simple test statistic and thus has a low implementation complexity. Secondly, SSCT is a sequential detection algorithm and is capable of achieving performance comparable to fixed sample size detection algorithms such as energy detection but with much reduced sensing time. Thirdly, SSCT is essentially a non-coherent detection algorithm in the sense that it does not require any deterministic knowledge of the primary signals. Lastly, SSCT is able to strike a desirable trade-off between sensing performance and sensing time particularly in the signal-to-noise ratio mismatched case. To evaluate sensing performance, we derive the exact false-alarm probability for SSCT, and develop numerical integration algorithms to compute misdetection probability and the average sample number. We further demonstrate the performance of SSCT with several numerical examples.

The mobility status of self and surrounding vehicles provides important information to various tasks in autonomous driving (AD) and intelligent transportation system (ITS). Accordingly, a precise, stable, and robust mobility tracking framework is essential. Compared with self-tracking that relies only on mobility observations from onboard sensors [e.g., global positioning system (GPS), inertial measurement unit (IMU), and camera], cooperative tracking markedly increases the precision and reliability of the mobility information by integrating observations from roadside units (RSUs) and nearby vehicles through vehicle-to-everything (V2X) communications in the Internet of Vehicles (IoV). Nevertheless, cooperative tracking can be quite vulnerable if there are malicious users sending bogus observations in the cooperative network. In this article, we present a malicious user detection framework, which includes two sequential detection algorithms and a secure mobility data exchange and fusion model to detect and remove bogus mobility information and integrate proposed detection algorithms with previous data fusion algorithms, which secures the cooperative mobility tracking in AD, ITS. Simulations validate the effectiveness and robustness of the proposed framework under different types of attacks.

In this paper, we develop an online sequential learning algorithm for single hidden layer feedforward networks (SLFNs) with additive or radial basis function (RBF) hidden nodes in a unified framework. The algorithm is referred to as online sequential extreme learning machine (OS-ELM) and can learn data one-by-one or chunk-by-chunk (a block of data) with fixed or varying chunk size. The activation functions for additive nodes in OS-ELM can be any bounded nonconstant piecewise continuous functions and the activation functions for RBF nodes can be any integrable piecewise continuous functions. In OS-ELM, the parameters of hidden nodes (the input weights and biases of additive nodes or the centers and impact factors of RBF nodes) are randomly selected and the output weights are analytically determined based on the sequentially arriving data. The algorithm uses the ideas of ELM of Huang et al. developed for batch learning which has been shown to be extremely fast with generalization performance better than other batch training methods. Apart from selecting the number of hidden nodes, no other control parameters have to be manually chosen. Detailed performance comparison of OS-ELM is done with other popular sequential learning algorithms on benchmark problems drawn from the regression, classification and time series prediction areas. The results show that the OS-ELM is faster than the other sequential algorithms and produces better generalization performance.