A GMM/HMM model for reconstruction of missing speech spectral components for continuous speech recognition

This paper presents a method for reconstructing unreliable spectral components of speech signals using the statistical distributions of the clean components. Our goal is to model the temporal patterns in speech signal and take advantage of correlations between speech features in both time and frequency domain simultaneously. In this approach, a hidden Markov model (HMM) is first trained on clean speech data to model the temporal patterns which appear in the sequences of the spectral components. Using this model and according to the probabilities of occurring noisy spectral component at each states, a probability distributions for noisy components are estimated. Then, by applying maximum a posteriori (MAP) estimation on the mentioned distributions, the final estimations of the unreliable spectral components are obtained. The proposed method is compared to a common missing feature method which is based on the probabilistic clustering of the feature vectors and also to a state of the art method based on sparse reconstruction. The experimental results exhibits significant improvement in recognition accuracy over a noise polluted Persian corpus.     

Optimal Deep Learning Based Intruder Identification in Industrial Internet of Things Environment

With the increased advancements of smart industries, cybersecurity has become a vital growth factor in the success of industrial transformation. The Industrial Internet of Things (IIoT) or Industry 4.0 has revolutionized the concepts of manufacturing and production altogether. In industry 4.0, powerful Intrusion Detection Systems (IDS) play a significant role in ensuring network security. Though various intrusion detection techniques have been developed so far, it is challenging to protect the intricate data of networks. This is because conventional Machine Learning (ML) approaches are inadequate and insufficient to address the demands of dynamic IIoT networks. Further, the existing Deep Learning (DL) can be employed to identify anonymous intrusions. Therefore, the current study proposes a Hunger Games Search Optimization with Deep Learning-Driven Intrusion Detection (HGSODL-ID) model for the IIoT environment. The presented HGSODL-ID model exploits the linear normalization approach to transform the input data into a useful format. The HGSO algorithm is employed for Feature Selection (HGSO-FS) to reduce the curse of dimensionality. Moreover, Sparrow Search Optimization (SSO) is utilized with a Graph Convolutional Network (GCN) to classify and identify intrusions in the network. Finally, the SSO technique is exploited to fine-tune the hyper-parameters involved in the GCN model. The proposed HGSODL-ID model was experimentally validated using a benchmark dataset, and the results confirmed the superiority of the proposed HGSODL-ID method over recent approaches.     

F-LEACH: a fuzzy-based data aggregation scheme for healthcare IoT systems

The Journal of Supercomputing - Internet of Things (IoT) is an emerging paradigm that consists of numerous connected and interrelated devices with embedded sensors, exchanging data with each other...     

Trends in intelligent manufacturing research: a keyword co-occurrence network based review

Journal of Intelligent Manufacturing - In recent years, driven by Industry 4.0 wave, academic research has focused on the science, engineering, and enabling technologies for intelligent and cyber...     

ETAS: predictive scheduling of functions on worker nodes of Apache OpenWhisk platform

The Journal of Supercomputing - Fast execution of functions is an inevitable challenge in the serverless computing landscape. Inefficient dispatching, fluctuations in invocation rates, burstiness...     

The Effect of Dilute Charged Impurity on the Electronic Heat Capacity and Magnetic Susceptibility of Ferromagnetic MoS<Subscript>2</Subscript>

In this paper, the effects of dilute charged impurity doping on electronic heat capacity (EHC) and magnetic susceptibility (MS) of a two-dimensional material ferromagnetic gapped graphene-like, MoS₂, are investigated within the Green’s function approach by using the Kane-Mele Hamiltonian and self-consistent Born approximation (SCBA) at Dirac points. Our findings show that there is a critical impurity concentration (IC) and scattering strength (ISS) for each valley in EHC and MS curves. Also, we have found that the spin band gap decreases with impurity only for valley K, ↑ and \(K^{\prime }, \downarrow \) due to the existence of inversion symmetry between valleys. On the other hand, a magnetic phase transition from ferromagnetic to antiferromagnetic and paramagnetic has been observed. The increase of scattering rate of carriers in the presence of impurity is the main reason of these behaviors.     

Discrete modeling of sand–tire mixture considering grain-scale deformability

Mixing sand or soil with small pieces of tire is common practice in civil engineering applications. Although the properties of the soil are changed, it is environmentally friendly and sometimes economical. Nevertheless, the mechanical behavior of such mixtures is still not fully understood and more numerical investigations are required. This paper presents a novel approach for the modeling of sand–tire mixtures based on the discrete element method. The sand grains are represented by rigid agglomerates whereas the tire grains are represented by deformable agglomerates. The approach considers both grain shape and deformability. The micromechanical parameters of the contact law are calibrated based on experimental results from the literature. The effects of tire content and confining pressure on the stress–strain response are investigated in detail by performing numerical triaxial compression tests. The main results indicate that both strength and stiffness of the samples decrease with increasing tire content. A tire contact of 40% is identified as the boundary between rubber-like and sand-like behavior.     

Classification of Electroencephalogram Signals Using LSTM and SVM Based on Fast Walsh-Hadamard Transform

Classification of electroencephalogram (EEG) signals for humans can be achieved via artificial intelligence (AI) techniques. Especially, the EEG signals associated with seizure epilepsy can be detected to distinguish between epileptic and non-epileptic regions. From this perspective, an automated AI technique with a digital processing method can be used to improve these signals. This paper proposes two classifiers: long short-term memory (LSTM) and support vector machine (SVM) for the classification of seizure and non-seizure EEG signals. These classifiers are applied to a public dataset, namely the University of Bonn, which consists of 2 classes –seizure and non-seizure. In addition, a fast Walsh-Hadamard Transform (FWHT) technique is implemented to analyze the EEG signals within the recurrence space of the brain. Thus, Hadamard coefficients of the EEG signals are obtained via the FWHT. Moreover, the FWHT is contributed to generate an efficient derivation of seizure EEG recordings from non-seizure EEG recordings. Also, a k-fold cross-validation technique is applied to validate the performance of the proposed classifiers. The LSTM classifier provides the best performance, with a testing accuracy of 99.00%. The training and testing loss rates for the LSTM are 0.0029 and 0.0602, respectively, while the weighted average precision, recall, and F1-score for the LSTM are 99.00%. The results of the SVM classifier in terms of accuracy, sensitivity, and specificity reached 91%, 93.52%, and 91.3%, respectively. The computational time consumed for the training of the LSTM and SVM is 2000 and 2500 s, respectively. The results show that the LSTM classifier provides better performance than SVM in the classification of EEG signals. Eventually, the proposed classifiers provide high classification accuracy compared to previously published classifiers.     

Efficient utilization of elliptic curve cryptosystem for hierarchical access control

The elliptic curve cryptosystem (ECC) has recently received significant attention by researchers due to its high performance, low computational cost, and small key size. In this paper, an efficient key management and derivation scheme based on ECC is proposed to solve dynamic access problems in a user hierarchy. Compared to previous ECC based works, the proposed method does not require constructing interpolate polynomials, therefore, the computational complexity of key generation and key derivation is significantly reduced. At the same time, time complexity of adding/deleting security classes, modifying their relationships, and changing of secret keys is decreased in the proposed method.