In this paper, we study physical layer security in an underlay cognitive radio (CR) network. We consider the problem of secure communication between a secondary transmitter-receiver pair presence randomly distributed eavesdroppers under interference constraint set by primary user. For different channel knowledge assumptions at transmitter, design four transmission protocols to achieve CR give comprehensive performance analysis for each protocol terms delay, security, reliability, and overall...

Bearings are crucial components of modern mechanical equipment and their failure can lead to downtime, economic losses, potential threats personal safety. Therefore, it is essential carry out fault diagnosis bearings. In this paper, Multichannel Signal Transformer (MST) model employed for diagnosing bearing faults. Initially, a single vibration signal two motor current signals collected combined into multichannel signal. This then used train test the MST model. The performance subsequently...

Rolling bearings are the core devices in industrial field, widely used machinery production and automation industry. Therefore, it is necessary to detect fault of rolling bearing. In this paper, an improved VGG model based on Channel Attention mechanism (SENet) proposed extract spatial features multi-scale after signal preprocessing. The BiGRU network Global optimization time domain pretreatment. Excellent results obtained by parallel fusion for detection.

With the rapid development of information technology and increasingly wide application Internet in field education, traditional teaching mode is facing a profound change. Blended teaching, as new integrating classroom online learning, has gradually become an important direction educational reform. The purpose this paper to discuss necessity feasibility blended reform higher mathematics courses era, analyse difficulties existing current put forward corresponding strategies, with view...

This paper investigates the secure transmission in simultaneous wireless information and power transfer (SWIPT) system, where a source node communicates with wireless‐powered full duplex destination presence of passive eavesdropping node. To take advantage benefits from SWIPT techniques, we propose two‐phase time‐sharing protocol: harvests energy form first phase, while information‐bearing signal is transmitted under protection artificial noise sent second phase. The transmit converted...

In order to get more effective expression features, this paper proposes an approach based on Gabor feature and Histogram of Oriented Gradients (HOG). Wavelet filter is first used as preprocessing stage for extraction. Handing the characteristics with a large number dimensions, binary encoding (BC) applied dimensionality reduction. Dimensionality vector reduced by using HOG algorithm. Experiments were performed Cohn-Kanade facial database support machine classifier classification. We obtained...

The secret key generation rate (SKGR) of current physical layer security methods is not high enough for practical requirements. A novel SKG method based on the cross multiplication (XM) two-way random signals proposed TDD-MIMO system. principle concealed beneath XM found that operation only derives common source both legitimate partners, but also employs new generated cross-terms due to bilinear provide more measurements and keys accordingly, which essentially superior existing paired (PM)...

Vehicle logo acquisition or recognition has been a popular study field in intelligent traffic system for the latest decade. In this paper, vehicle method based on CNN (Convolutional Neural Network) is introduced. experiment, two classification methods with different feature extraction were applied. However, these utilized same SVM classifier procedure. The experiment uses input images height of 140 pixels and width 100 pixels. samples include 26 classes, which are derived from 16 brands....

A simple and efficient one-pot protocol has been developed by an aza-Diels–Alder mechanism for the stereoselective synthesis of novel fused multicyclic iminosugars with structural diversity.

This paper adopts a wrapper method to find subset of features that are most relevant the classification task. The approach utilizes an improved estimation conditional mutual information which is used as independent measure for feature ranking in local search operations. Meanwhile, between predictive labels trained classifier and true classes fitness function global best features. Thus, searches consist hybrid genetic algorithm selection. Experimental results demonstrate both parsimonious...

Feature extraction is the important prerequisite of classifying text effectively and automatically. TF·IDF widely used to express feature weight. But it has some problems. can't reflect distribution terms in text, then importance degree difference between categories. This paper proposes a new weighting method-TF·IDF·C <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i</sub> which weight Ci added differences classes on base original TF·IDF. After...

This paper investigates the physical-layer security of uplink non-orthogonal multiple access (NOMA) in cellular Internet Things (IoT) with invoking stochastic geometry. Poisson cluster process-based model is applied to characterize NOMA transmission scenario, where IoT terminals are located around serving base station. Considering severe interference brought by a large number terminals, inter-cell also taken into account. To enhance limited overhead increment at stations not only receive...

Analytic ally studying the overall outage performance of two-way amplify-and-forward (AF) relaying systems has always been a difficult task. In this paper we study probability (OOP) for system. The channels are independent but not identically Nakagami-m fading distributed. We obtain tight closed-form OOP approximated expression Our results indicate that approximation can march simulation perfectly in large signal-to-noise ratios (SNRs). theoretical analysis be further applied to...

Accurate fault diagnosis for large machines is very important its economic meaning. In essence, a pattern classification and recognition problem of judging whether the operation status system normal or not. Support vector (SVMs), well motivated theoretically, have been introduced as effective methods solving problems. However, generalization performance SVMs sometimes far from expected level due to their limitations in practical applications. Ensemble learning provides promising way such...

When the eavesdropping channel has a certain correlation with legitimate channel, physical layer key is susceptible to eavesdroppers. To ensure security of secret key, we propose two-layer secure (TLS) quantization algorithm in this paper. First, model between distance and derive joint probability density function phase capacity. Then, describe detail TLS prove validity our algorithm. Finally, evaluate performances using two parameters-key rate bit disagreement rate. The Simulation results...

The Pearl River Delta, one of the main regions China's export-oriented economy, has benefited from its traditional economic structure for three decades which in turn appeared to hinder further development recently. To advance industrial restructuring and upgrading, governments have promulgated policies appropriated special funds optimize Delta. Dongguan is a famous manufacturing city China. Taking as case, applying method Difference-in-Difference (DID), using time-series data 1997 2014, this...

To solve the problem that key is susceptible to eavesdroppers in correlated channel scenarios, this paper, we propose a native security scheme based on physical layer chain keys, which can be used for encryption and authentication. First, generated by performing XOR operation updated previous key, where obtained current state information. The proposed uses as authentication root, relationship continuously transferred grows. Then, CRC realize identity message verification simultaneously....

Co-prime linear arrays (CLAs) provide an additional degree of freedom (DOF) with a limited number physical sensors, and thus help to improve the resolution direction arrival (DOA) estimation algorithms. However, DOF traditional CLA is restrained by structure array, which cannot be adjusted after deployment. In this paper, we propose DOA algorithm for reconfigurable intelligent surface co-prime array (RIS CLA) based on multiple signal classification approach. Specifically, RIS first...

Principal component analysis (PCA) is widely used in fault diagnosis. Because the traditional data preprocessing method ignores correlation between different variables system, feature extraction not accurate. In order to solve it, this paper proposes a kind of based on Gap metric improve performance PCA For types faults, original dataset transformation through can reflect system high-dimensional space, so as model more accurately. Finally, feasibility and effectiveness proposed are verified...