Search

Investigation of the behaviour and complexity of hard-to-measure parameter time series is not explored to a greater extent before modelling a wastewater treatment facility (WWTF). In this context, A dynamic non-linear chaotic approach, namely the False nearest neighbour (FNN) algorithm, is employed for the first time to investigate the influent and effluent quality parameters of a WWTF. The primary objective of this research is to analyze the parameters of a WWTF located in India for its behaviour and complexity using the FNN algorithm. The autocorrelation function and average mutual information time lags are used as the delay time (τ) in the algorithm for phase space reconstruction and further FNN analysis. The optimum embedding dimensions (mopt) from FNN plots indicate the complexity or number of optimum variables required to model the time series. For influent and effluent parameters, the mopt values fall within a range of 4–15 and 4–17, respectively, and the τ value influences this range. Wastewater time series behaviour differs, such as pure stochastic or chaotic or chaotic series with noise, which is highly dependent on τ. The future scope of the study involves integrating the retrieved behaviour and complexity into state-of-the-art artificial intelligence or data-driven techniques to forecast hard-to-measure parameters. [ABSTRACT FROM AUTHOR]

Security in mobile ad-hoc networks (MANETs) is a key issue to be addressed. There are two separate approaches for security in MANETs viz., continuous authentication and intrusion detection. Previous studies in this regard involved a study of the two classes of issues separately. In this work these two classes of security approaches are integrated and combined into a single console. Continuous authentication acts as the first line of defense while intrusion detection system acts as the second. In this framework, biometric authentication is used for continuous authentication. The intrusion detection system has a detection engine and a response engine. The detection engine detects the malicious intruder and the response engine completely eliminates the above detected intruder from the network. The analysis of the security issues is simulated to produce the necessary results. [ABSTRACT FROM AUTHOR]

ABSTRACT This article proposes a novel and efficient methodology for the detection of Glioblastoma tumor in brain MRI images. The proposed method consists of the following stages as preprocessing, Non-subsampled Contourlet transform (NSCT), feature extraction and Adaptive neuro fuzzy inference system classification. Euclidean direction algorithm is used to remove the impulse noise from the brain image during image acquisition process. NSCT decomposes the denoised brain image into approximation bands and high frequency bands. The features mean, standard deviation and energy are computed for the extracted coefficients and given to the input of the classifier. The classifier classifies the brain MRI image into normal or Glioblastoma tumor image based on the feature set. The proposed system achieves 99.8% sensitivity, 99.7% specificity, and 99.8% accuracy with respect to the ground truth images available in the dataset. [ABSTRACT FROM AUTHOR]