Search

Phishing is one of the most harmful social engineering techniques to subdue end users where threat actors find a chance to gain access to critical information systems. A common approach in phishing is through the use of e-mail communication with an embedded hyperlink. The detection and mitigation of phishing attacks are a grand challenge due to the complexity of current phishing attacks. Existing techniques are often too time consuming to be used in the real world in terms of detection and mitigation time. Likewise, they employ static detection rules that are not effective in the real world due to the dynamics of phishing attacks. In this paper, we present PhishLimiter, a new detection and mitigation approach, where we first propose a new technique for deep packet inspection (DPI) and then leverage it with software-defined networking (SDN) to identify phishing activities through e-mail and web-based communication. The proposed DPI approach consists of two components: phishing signature classification and real-time DPI. Based on the programmability of SDN, we develop the store and forward mode and the forward and inspect mode to the direct network traffic by using an artificial neural network model to classify phishing attack signatures and design the real-time DPI so that PhishLimiter can flexibly address the dynamics of phishing attacks in the real world. PhishLimiter also provides better network traffic management for containing phishing attacks since it has the global view of a network through SDN. Furthermore, we evaluate PhishLimiter using a real-world testbed environment and data sets consisting of real-world e-mail with embedded links. Our extensive experimental study shows that PhishLimiter provides an effective and efficient solution to deter malicious activities.

Flying ad hoc networks (FANETs) that consist of multiple unmanned aerial vehicles (UAVs) are promising technologies for future networked systems due to the versatility of UAVs. One of the most distinguishing features of FANET is frequent and rapid topological fluctuations due to the high-mobility of UAVs. Hence, the topology management adapting to the movements of UAVs is one of the most critical issues in FANET. In this paper, we study a FANET topology management problem that optimizes the locations and movements of UAVs to maximize the network performance, adapting to the topological changes while UAVs carry out their missions. When formulating the problem, we take into account the routing protocol as an arbitrary function since the network performance is inseparably linked with the routing protocol in use. We first develop two algorithms. One is the topology construction algorithm, which constructs a FANET topology from the scratch without any given initial topology, based on particle swarm optimization. The other is the topology adjustment algorithm, which incrementally adjusts the FANET topology adapting to the movements of UAVs with low-computational costs, based on gradient descent. Then, by defining a logical distance (the so-called topology edit distance) that measures the degree of changes in FANET topology, we develop an integrated topology management algorithm that contains the topology construction and adjustment algorithms. The simulation results show that our algorithm achieves a good network performance with low computational overhead, which is one of the most essential virtues in FANETs with rapidly varying topology.

Fault diagnosis technique is the core of Prognostics and Health Management (PHM) system, which plays a crucial role in the intelligent operation and maintenance of various rotating machineries. In this paper, we present a novel sparse representation classification framework with structured dictionary design strategy (SRC-SDD) for intelligent fault diagnosis of rotating machineries. The proposed SRC-SDD method consists of two stages, i.e., the structured dictionary design stage and the sparsity-based intelligent diagnosis stage. In the first stage, the novelty of SRC-SDD lies in the overlapping segmentation strategy for structured dictionary design, which leverages the structured prior knowledge of rotating machinery vibration signals, namely, the periodic self-similarity and shift-invariance properties. In the second stage, SRC-SDD achieves fault recognitions of testing samples using a sparsity-based diagnosis strategy based on the minimum sparse reconstruction error. The proposed structured dictionary design strategy can enhance the representation power of dictionaries and thus promote the recognition performance of the sparsity-based diagnosis strategy. Finally, the effectiveness of SRC-SDD has been validated on the gearbox fault dataset from IEEE PHM society. The diagnosis results show that SRC-SDD achieves the excellent recognition accuracy of 100% for predicting six different gearbox health states. Further, the comparative studies with three conventional SRC methods prove the superiority of SRC-SDD in terms of both the recognition performance and computation efficiency.

Non-orthogonal multiple access (NOMA) is a promising spectrally-efficient technology to satisfy the massive data requirement of the next-generation wireless communication networks. In this paper, we consider cooperative non-orthogonal multiple access (CNOMA) networks with relaying user selection, where a two-stage user selection scheme is adopted to optimize the received signal-to-noise ratio (SNR) of the weak user. We derive analytical expressions of the outage probability of the CNOMA users under two different types of channel estimation errors. Afterwards, we investigate the asymptotic characteristics the outage probability of the CNOMA systems in the high SNR regimes. Our results show that for the case of constant variance of the channel estimation error, the outage probability of each user is saturated and cannot be improved by increasing the transmit power of the access point. In contrast, there is no performance bottleneck of the outage probability for all the users when the variance of the channel estimation errors decease with SNR, whilst fully diversity gain is achieved by the weak user.

This paper proposes a robust position control scheme for a quadrotor UAV system under uncertainties. The proposed control algorithms combine integral sliding mode and backstepping sliding mode controllers in a double-loop control structure (i.e., inner-outer loop control). The design of the proposed controller is divided into two subcontrollers, namely, attitude and position controllers for the quadrotor. In this work, a nonsimplified six-degree-of-freedom quadrotor model is first established in the presence of disturbances. Afterward, we develop a robust backstepping sliding mode controller for the attitude control of the quadrotor. Next, a robust integral sliding mode controller is designed for the outer loop of the quadrotor to ensure the position trajectory tracking capability in the presence of disturbances. The stability and performance of the quadrotor is thoroughly investigated using Lyapunov stability analysis. Numerical simulations demonstrate the effectiveness of the developed solutions for a quadrotor.

This paper is concerned with the problems of finite-time stochastic $H_\infty $ control for singular Ito Markovian jump systems with $(x,v)$ -dependent noise and generally uncertain transition rates (GUTRs). Based on two equivalent sets, a new criterion, ensuring the considered systems with completely known TRs to be finite-time stochastically bounded with $H_\infty $ performance, is first established, which is less conservative than the existing one. Then, the obtained results are extended to the case of GUTRs. To overcome the nonlinear difficulty resulting in the Young inequality for designing a convex controller, an approach called the association of free variables and slack variables are used. The state feedback controller and observer-based controller are respectively designed such that the corresponding closed-loop systems with GUTRs are finite-time stochastically bounded while achieving $H_\infty $ performance. Finally, the numerical examples are addressed to illustrate the effectiveness and efficiency of our obtained results.

Internet traffic volume is continuing to increase rapidly. Named data networking (NDN) has been introduced to support this Internet traffic growth through caching contents close to consumers. While caching in NDN is beneficial to both Internet service providers (ISPs) and content providers (CPs), ISPs serve cached contents independently without any coordination with CPs. By authorizing the ISPs to cache and distribute the contents accessible on payments, it becomes impractical for CPs to control content access and payments. In this paper, we address these challenges by proposing a joint incentive mechanism and a price-based cache replacement (PBCR) policy for paid content in NDN that improves the ISP's and CPs' profits. We use an auction theory, where the ISP earns profits from caching by alleviating traffic load on transit links and participating in contents selling. Therefore, before the ISP starts selling cached contents, it needs to cache them first. Furthermore, the ISP cache capacity is limited; therefore, we propose PBCR, where the PBCR triggers the content that needs to be replaced when the cache storage is full based on both content price and link cost. The simulation results show that our proposal increases the profits of all the network players involved in paid content caching and improves cache hit ratio. Published version

With the development of machine learning for decades, there are still many problems unsolved, such as image recognition and location detection, image classification, image generation, speech recognition, natural language processing and so on. In the field of deep learning research, the research on image classification has always been the most basic, traditional and urgent research direction. At the same time, computer intelligent image recognition technology is also conducive to gradually better respond to the development of international indicators, and promote the development and progress of various fields. Therefore, image processing technology based on machine learning has been widely used in feature image, classification, segmentation and recognition, and is a hot spot in various fields. However, due to the complexity of video images and the distribution of objects in different application backgrounds, the classification accuracy becomes important and difficult. In the paper transportation industry, image recognition technology is applied to license plate recognition to extract license plate from complex background, segment license plate characters and recognize characters, and construct a machine learning non license plate automatic generation algorithm, which may improve the efficiency of non license plate recognition. The diversity and high generation speed of license plate training sample set can achieve the purpose of effectively training strong classifier. By using genetic algorithm to optimize BP neural network to classify license plate information, the anti-interference ability and license plate recognition accuracy are improved to a certain extent.

Analysis of non-stationary signals in a noisy environment is a challenging research topic in many fields often requiring simultaneous signal decomposition in the time and frequency domain. This paper proposes a method for the classification of noisy non- stationary time- series signals based on Cohen’s class of their time- frequency representations (TFRs) and deep learning algorithms. We demonstrated the proposed approach on the example of detecting gravitational-wave (GW) signals in intensive real- life, non- stationary, non-white, and non-Gaussian noise. For this purpose, we prepared a dataset based on the actual data from the Laser Interferometer Gravitational-Wave Observatory (LIGO) detector and the synthetic GW signals obtained by realistic simulations. Next, 12 different TFRs from Cohen’s class were calculated from the original noisy time-series data and used to train three state-of- the-art convolutional neural network (CNN) architectures: ResNet-101, Xception, and EfficientNet. The obtained classification results are compared to those achieved by the base model trained on the original time series. Analysis of the results suggests that the proposed approach combining deep CNN architectures with Cohen’s class TFRs yields high values of performance metrics and significantly improves the classification performance compared to the base model. The TFR-CNN models achieve the values of the classification accuracy of up to 97.10%, the area under the receiver operating characteristic curve (ROC AUC) of up to 0.9885, the recall of up to 95.87%, the precision of up to 99.51%, the F1 score of up to 97.03%, and the area under the precision-recall curve (PR AUC) of up to 0.9920. This classification performance is obtained on the dataset in which the signal- to- noise ratio (SNR) values of the raw, noisy time- series signals range from -123.46 to -2.27 dB. Therefore, this study suggests that using alternative TFRs of Cohen’s class can improve the deep learning-based detection of non-stationary GW signals in an intensive noise environment. Moreover, the proposed approach can also be a viable solution for deep learning-based analysis of numerous other noisy non-stationary signals in different practical applications.

Power inversion algorithms (PI) are widely used by ant-jamming receivers in harsh environment, but it also induces pseudo-range measurement error which is fatal to high precision receivers. In this paper, it proposes a more accurate method based on the S Curve Bias(SCB) to evaluate the pseudo-range bias which can actually affect the positioning results, as it excludes a fixed time delay contained in the directly differing method. When using this method to assess the bias, it is found that pseudo-range errors can be influenced by many factors including the type of interference. Simulation as well as experiment based on a 4-element array antenna are designed and carried out. The simulation has shown that both SOP-PI and STAP-PI will bring bias to pseudo-range measurement. Specifically, the maximum bias caused by SOP-PI is 1m while the bias caused by STAP-PI is up to 10m. Moreover, comparing to the method which directly differs the pseudo-range of output signal and the reference signal,the method proposed in this paper has better performance. Furthermore, the experimental results verify the effectiveness of the proposed evaluating method.

Estimating semantic correspondence between pairs of images can be challenging as a result of intra-class variation, background clutter, and repetitive patterns. This paper proposes a convolutional neural network (CNN) that attempts to learn rich semantic representations that contain the global semantic context to enable robust semantic correspondence estimation against intra-class variation and repetitive patterns. We introduce a global context fused feature representation that efficiently employs the global semantic context in estimating semantic correspondence as well as a semi-global self-similarity feature to reduce background clutter-induced distraction in capturing the global semantic context. The proposed network is trained in an end-to-end manner using a weakly supervised loss, which requires a weak level of supervision involving annotation on image pairs. This weakly supervised loss is supplemented with a historical averaging loss to effectively train the network. Our approach decreases running time by a factor of more than four and reduces the training memory requirement by a factor of three and produces competitive or superior results relative to previous approaches on the PF-PASCAL, PF-WILLOW, and TSS benchmarks.

Cricothyrotomy procedures, involving risky incisions on the neck skin and internal membranes, require rigorous training. In this paper, a novel incision sensor measuring the incision path in a cricothyrotomy training simulation is proposed. The sensor provides quantitative feedback to trainees on their incision practice, enhancing the effectiveness of the simulation. The sensor measures the electric potential, which decreases monotonically along the direction of current flow on the conductive material at the incision point, and converts it to coordinate values, based on the relationship between the electric potential and the position. The sensor comprises three layers of conductive tape, which are electrically isolated by two dielectric layers, and is fabricated as a thin film. The first two conductive layers (driving layers) are alternately energized to create distributions of electric potential in the $x$ or $y$ directions across the sensor plane. The third conductive layer (sensing layer), placed under the driving layers, transfers the electric potential to the output channel of the sensor at the point where a metal blade creates a short circuit between the energized driving layer and the sensing layer. The alternating measurements are converted to $x$ and $y$ coordinates of the incision position. The experiments for characterization and performance validation were performed using sensor prototypes fabricated with the proposed design and fabrication procedures. The experimental results show that the proposed sensor facilitates the measurement of the incision paths aligned with and diagonal to the $x$ and $y$ axes within root mean square errors of 0.98 and 1.03 mm, respectively.

For downlink multiple-input multiple-output-sparse code multiple access (MIMO-SCMA) system, the convergence reliability of the detected codeword in each decision process is unsatisfactory at the message passing algorithm receiver. To address this problem, a downlink MIMO-mixed sparse code multiple access (MIMO-MSCMA) system is proposed in this paper. For each user of the downlink MIMO-MSCMA, the transmitted codewords at different antennas come from different codebooks of MSCMA, respectively. On the other side, a novel basic complex multi-dimension constellation design is proposed in the MSCMA codebook design. Simulation results show that the bit error rate performance of the downlink MIMO-MSCMA outperforms than that of downlink MIMO-SCMA.

In cooperative relaying, intermediate stations are required to enhance the end-to-end transmission performance. The performance of the cooperative relaying scheme has been investigated theoretically and via computer simulations. However, cooperative relaying using transmit diversity techniques in actual environments has not been investigated thus far. This paper presents an experimental system for distributed cooperative relaying using space-time block code and evaluations of its transmission performances in real propagation channels. To this end, four wireless stations-specifically, one source, two relays, and one destination-were developed using analog transceivers and field-programmable gate arrays for real-time digital signal processing. Sample timing and frequency synchronizations among the four wireless stations were established by using the received signals as a reference. The end-to-end error performance of distributed cooperative relaying was compared to those of noncooperative relaying schemes, and the performances of three relaying schemes were evaluated quasisimultaneously in terms of their cumulative distribution functions of the bit-error ratios (BERs). The experimental results indicated that the BER performance of the two-hop distributed cooperative relaying scheme was substantially superior to those of noncooperative two-hop relaying schemes, including a route diversity scheme.

Differential privacy is the state-of-the-art for preserving privacy and differential privacy mechanism based on Laplace distribution with mean 0 is common practice. However, privacy budget is exhausted so quick that the number of queries is not enough. In this paper, a differential privacy mechanism is proposed to optimize the number of queries for application scenario of multiple users. We isolate different users by assigning various noise distribution with non-zero mean to different users. First, in terms of privacy guarantee, the proposed mechanism is better than common practice. Second, for the utility aspect, the accuracy of proposed mechanism is analyzed from the view of data distribution's distortion and the view of noise's absolute value.

Ambitious quality of experience expectations from 5G mobile cellular networks have spurred the research towards ultra-dense heterogeneous networks (UDHNs). However, due to coverage limitations of millimeter wave cells and lack of coverage data in UDHNs, discovering coverage lapses in such 5G networks may become a major challenge. Recently, numerous studies have explored machine learning-based techniques to detect coverage holes and cell outages in legacy networks. Majority of these techniques are susceptible to noise in the coverage data and only characterize outages in the spatial domain. Thus, the temporal impact of an outage, i.e., the duration of its presence remains unidentified. In this paper, for the first time, we present an outage detection solution that characterizes outages in both space and time while also being robust to noise in the coverage data. We do so by employing entropy field decomposition (EFD) which is a combination of information field theory and entropy spectrum pathways theory. We demonstrate that compared to other techniques such as independent component analysis and k-means clustering, EFD returns accurate detection results for outage detection even in the presence of heavy shadowing in received signal strength data which makes it ideal for practical implementation in emerging mobile cellular networks.

Computational inference of aesthetics has recently become a hot topic due to its usefulness in widely applications such as evaluating image quality, retouching image and retrieving image. Owing to the subjectivity of this problem, there is no general framework to predict image aesthetics. In this paper, we propose a deep neural framework with visual attention module, self-generated global features and hybrid loss to address this problem. Specifically, the framework can be any state-of-the-art convolution classification network compatible with visual attention. Further, self-generated global feature compensates for the loss of global context information during training stage, and the hybrid loss guides the network to learn the similarity between the predicted aesthetic scores and the ground-truths through fusing soft-max-entropy and Earth Mover’s Distance(EMD). With the above-mentioned improvements, the proposed deep neural framework is capable of effectively predicting image aesthetics in an efficient way. In our experiments, we release a real-world aesthetic dataset that contains 1,800 2K photos labeled by several experienced photographers, and then provide a thorough ablation study of the design choices to better understand the superiority brought by each part of our framework, and design several comparisons with the state-of-the-art methods on a fraction of metrics. The experimental results on two datasets demonstrate that both accuracy and efficiency achieve favorably performance.

Recently, deep learning-based image compression has shown significant performance improvement in terms of coding efficiency and subjective quality. However, there has been relatively less effort on video compression based on deep neural networks. In this paper, we propose an end-to-end deep predictive video compression network, called DeepPVCnet, using mode-selective uni- and bi-directional predictions based on multi-frame hypothesis with a multi-scale structure and a temporal-context-adaptive entropy model. Our DeepPVCnet jointly compresses motion information and residual data that are generated from the multi-scale structure via the feature transformation layers. Recent deep learning-based video compression methods were proposed in a limited compression environment using only P-frame or B-frame. Learned from the lesson of the conventional video codecs, we firstly incorporate a mode-selective framework into our DeepPVCnet with uni- and bi-directional predictive modes in a rate-distortion minimization sense. Also, we propose a temporal-context-adaptive entropy model that utilizes the temporal context information of the reference frames for the current frame coding. The autoregressive entropy models for CNN-based image and video compression is difficult to compute with parallel processing. On the other hand, our temporal-context-adaptive entropy model utilizes temporally coherent context from the reference frames, so that the context information can be computed in parallel, which is computationally and architecturally advantageous. Extensive experiments show that our DeepPVCnet outperforms AVC/H.264, HEVC/H.265 and state-of-the-art methods in an MS-SSIM perspective.

This paper presents a new design for a wideband circularly polarized (CP) spiral antenna array fed by a differential aperture coupling. The differential excitation is realized by two vias located at the lateral sides of a transverse slot etched on the lower substrate. The proposed two-arm spiral antenna has a smaller size compared to conventional spiral antennas with several turns. The presented single radiating element provides the impedance bandwidth of ( $|\text{S}11| dB) 40.4% and 3-dB axial ratio (AR) of 23.2% with a peak RHCP gain of 7.2 dBic over the operating frequency band. To further enhance the AR bandwidth and the maximum gain, the presented CP spiral element is employed in a $2\times 2$ sequential feeding arrangement. The sequential feeding network is implemented using the low-loss printed gap waveguide (PGW) line, which is compatible with low-cost printed circuit board (PCB) technology. A prototype of the proposed $2\times 2$ antenna array is fabricated and measured experimentally. Measured results reveal that the designed antenna array has a wide impedance bandwidth of over 48.5%, the 3-dB AR bandwidth of 34.2%, and the 3-dB gain bandwidth of 32.4% with a maximum gain of 11.74 dBic and excellent radiation efficiency of 85% over the entire operating frequency band.

In this paper, we apply polar codes to a modified version of OFDM systems with index modulation which is called OFDM with in-phase/quadrature index modulation (OFDM-I/Q-IM). We provide general design guidelines for the proposed polar-coded OFDM-I/Q-IM systems. In the proposed system, at the transmitter, we employ a random frozen bits appending scheme which not only makes the polar code compatible with OFDM-I/Q-IM but also improves the bit error rate (BER) performance of the system. Furthermore, at the receiver, it is shown that the a posteriori information for each index provided by the index detector is essential for the iterative decoding of polar codes by the belief propagation (BP) algorithm. Simulation results show that the proposed polar-coded OFDM-I/Q-IM system outperforms its OFDM counterpart in terms of BER performance.

This paper studies the problem of $H_{\infty }$ filtering for nonlinear Markov jump systems based on Takagi-Sugeno model. Firstly, we propose a hybrid event-triggered mechanism with an adjustable threshold, which not only helps to save more limited communication resources, but also excludes Zeno behavior while preserving the merits of continuous triggering. Secondly, given the threat of cyber-attacks to network security, a stochastic variable is introduced to describe the considered deception attacks in filter design. Thirdly, a less restrictive Lyapunov-Krasovskii functional (LKF), which is not required to be continuous and positive definite in a triggering interval, is constructed to establish sufficient condition on the exponential mean-square stability for the filtering error system with a weighted $H_{\infty }$ performance. Meanwhile, co-design of the desired filter and event-triggered mechanism is achieved. Finally, a tunnel diode circuit system is provided to illustrate the effectiveness and advantage of the obtained results.

Reinforcement learning (RL) is an unsupervised learning technique used in many real-time applications. The essence of RL is a decision-making problem. In RL, the agent constantly interacts with the environment and selects the next action according to previous feedback in terms of reward. In this paper, RL trains Software-Defined Wireless Sensor Networks (SDWSNs) controller to optimize the routing paths. We combine RL and SDN, where RL is applied to the SDN controller to generate the routing tables. We also propose four different reward functions for optimization of the network performance. RL-based SDWSN improves network performance by 23% to 30% in terms of lifetime compared with RL-based routing techniques. RL-based SDWSN performs well because it can intelligently learn the routing path at the controller. In addition, it has a faster network convergence rate than RL-based WSN.

Because the existing finger vein segmentation networks are too large and not suitable for implementation in mobile terminals, the reduction of the parameters of the lightweight network leads to the reduction of the segmentation index, and the long-running time of deep network on hardware platforms; this paper proposes a lightweight real-time segmentation method for finger veins based on embedded terminal technique. In the preprocessing stage of the algorithm, the data is greatly expanded by randomly selecting the center to obtain sub-blocks on each image of the training set. The network first uses deep separable convolution to greatly reduce the U-Net parameters of a basic network and introduces an attention module to reorder the features to improve network performance, followed by a preliminary lightweight network Dinty-NetV1. Second, the Ghost module is added to the deep separable convolution, and the feature map of the network part is obtained through a cheap operation so that the network is further compressed to obtain Dinty-NetV2. After adding channel shuffle, all the characteristic channels are evenly shuffled and reorganized to obtain Dinty-NetV3. Finally, a study of the filter norm yields the distribution characteristics of the finger vein picture features. By using the geometric median pruning method, the network models for each stage of the algorithm proposed in this paper achieved better segmentation performance and shorter split time after pruning. The overall Dinty-NetV3 model size is only less than 9% of the U-Net and Mult-Adds is less than 2% of the U-Net with the same structure. After testing on two-finger vein datasets SDU-FV and MMCUBV-6000, we confirm that the performance of Dinty-NetV3 surpasses all previously proposed classic compression model algorithms and it is not inferior to more complex and huge networks such as U-Net, DU-Net, and R2U-Net. The proposed algorithm has advantages in terms of time needed to train the network, and we verify its universality using NVIDIA's full range of embedded terminals.

The efficient and compact design of multilevel inverters (MLI) motivates in various applications such as solar PV and electric vehicles (EV). This paper proposes a 53-Level multilevel inverter topology based on a switched capacitor (SC) approach. The number of levels of MLI is designed based on the cascade connection of the number of SC cells. The SC cells are cascaded for implementing 17 and 33 levels of the output voltage. The proposed structure is straightforward and easy to implement for the higher levels. As the number of active switches is less, the driver circuits are reduced. This reduces the device count, cost, and size of the MLI. The solar panels, along with a perturb and observe (PO) algorithm, provide a stable DC voltage and is boosted over the DC link voltage using a single input and multi-output converter (SIMO). The proposed inverters are tested experimentally under dynamic load variations with sudden load disturbances. This represents an electric vehicle moving on various road conditions. A detailed comparison is made in terms of switches count, gate driver boards, sources count, the number of diodes and capacitor count, and component count factor. For the 17-level, 33-level, and 53-level MLI, simulation results are verified with experimental results, and total harmonic distortion (THD) is observed to be the same and is lower than 5% which is under IEEE standards. A hardware prototype is implemented in the laboratory and verified experimentally under dynamic load variations, whereas the simulations are done in MATLAB/Simulink.

In this paper, we introduce an ultra-high frequency radio frequency identification (UHF-RFID) mobile robot platform that is capable of performing fully autonomous inventory taking and stocktaking by providing three-dimensional (3D) product maps and thus making possible the concept of a smart warehouse. The proposed novel hardware architecture consists of an eight-channel UHF-RFID-listener for parallel signal phase recovery, including two different carrier leakage suppression circuits and a correlation decoder for each channel for the tag signal, which can handle a backscatter link frequency (BLF) deviation of up to ${\mathrm {22~\%}}$ to decode the tag data. The system also uses eight parallel channels for multiple-input multiple-output (MIMO) localization. For the system evaluation we labeled clothes stored in a warehouse with tags and generated their product map. The proposed localization algorithm is based on a synthetic aperture radar (SAR) MIMO approach that needs exact knowledge of the antenna positions and, therefore, of the driven trajectory. This position is provided by the robot, which takes advantage of a simultaneous localization and mapping (SLAM) algorithm, determining the position with ${\mathrm {1~cm}}$ accuracy while generating two-dimensional (2D) maps of the surroundings. We placed ten tags at known positions to assess the system’s performance and were able to locate these tags within a root mean square error (RMSE) of ${\mathrm {1.45~cm}}$ in 3D.

A new single-phase 15-level inverter with a reduced number of components for the Solar PV application is proposed in this paper. It is incorporate the proposed inverter with a boost converter to extract energy from the solar PV modules, the proposed inverter aids to generate fifteen stepped output voltage levels with lower THD. The proposed inverter can improve efficiency and reduce losses, cost, and complexity of the overall system. The conventional boost converter will boost the output voltage to a maximum voltage from Solar PV with MPPT(PO). The proposed inverter is tested experimentally using the dSPACE RTI 1104 controller along with MATLAB/Simulink. A detailed comparison with existing MLIs with the proposed inverter. The work presents experimental results not only to show its efficiency but also to the effectiveness under different circumstances of linear and non-linear loads. The inverter is stable during the non-linear loads and well suits for grid-connected systems.

Extreme weather like typhoon leaves significantly terrible influence on the stability of power grid in the coastal area. A high-voltage transmission line includes hundreds of towers and usually stretches hundreds of kilometers. It is not economical to install typhoon monitoring system throughout the transmission line. This paper proposes a typhoon inversion method (TIM) for transmission line which utilizes engineering wind field model and only requires monitoring data from a few towers. The developed method is called YanMeng wind field (YM) with directional mutation genetic algorithm (DMGA) for transmission lines, or YM-DMGA method. DMGA utilizes real-time monitoring data, and improves simulation accuracy of average wind speed by dynamically optimizing two critical parameters $B$ and $z_{0}$ in wind field model, which is a great improvement over the traditional YM model relying on historical data. The TIM is mainly composed of three parts, the meteorological monitoring system (MMS), the YM-DMGA model, and the software system. The TIM collects real-time measured wind data of particular monitoring stations through the MMS. Meanwhile, wind data of the whole transmission line is simulated using the YM-DMGA and displayed by the software system. Then the method proposed is verified to be effective in three aspects. First of all, the YM-DMGA method has great enhancement in accuracy, for the coefficient of determination $R_{square}$ increases from 0.811 to 0.986. Secondly, DMGA in the method has quicker convergent speed than typical GA, with fitness converged in the 269th epoch, earlier than that for typical GA, 656th epoch. Lastly, DMGA has better optimal fitness value of 0.6239, bigger than that of typical GA, 0.5891. At the end of the paper, an application of the method to an 110kV double-circuit transmission line is presented to reveal the risk under the impact of the super typhoon Rammasun.

Metro network is the important lifeline in modern cities. Its stability and reliability are critical for guaranteeing the urban residents' efficient commuting and continuous operation of urban functions. A series of disruptions show the vulnerability of metro network, and even the breakdown of a single node is sufficient to collapse the entire network due to cascading failure. However, the vulnerability assessments of metro network in the previous studies neglect the impact of cascading failure on the drop of network service performance. This paper measures the vulnerability of metro network by capturing the demand loss and travel delay under cascading failure. First, a load-capacity based model is developed to describe the dynamic process of cascading failures. Then, a weighted composite index composed of demand loss and travel time delay under cascading failure is used to measure metro network vulnerability. Finally, taking Shanghai metro network as an example, five attack scenarios are simulated to investigate cascading failure process and network vulnerability. The results reveal that cascading failures result in severe consequences in the metro network. The vulnerability of metro network without considering cascading failure is underestimated. The decrease in the tolerance parameter leads to the increase in metro network vulnerability. Random attack on one node is the most sensitive to the tolerance parameter. Node betweenness, cascading failure path and the initial load on the overloaded nodes also affect metro network vulnerability. This study provides a new perspective for understanding vulnerability of metro network, and also provides insights for improving operation reliability and stability of the network, as well as for designing emergency strategies to protect the network.

Some dynamical behaviors of fractional-order commutative quaternion-valued neural networks (FCQVNNs) are studied in this paper. First, because the commutative quaternion does not satisfy Schwartz triangle inequality, the FCQVNNs are divided into four real-valued neural networks (RVNNs) through quaternion commutative multiplication rules. Furthermore, several types of dynamical behaviors including global Mittag-Leffler stability, the boundedness with bounded disturbances, complete synchronization and quasi-synchronization of FCQVNNs are studied. Simultaneously, several conditions for these dynamical behaviors are driven by fractional-order Lyapunov direct method, some inequality techniques and fractional differential equation theory. At last, the effectiveness and feasibility of the obtained theoretical results are verified by several numerical simulation examples.

Wind power, photovoltaics and other new energy sources are connected to the grid on a large scale. The power electronic interface cannot provide inertia and damping support for the microgrid. The virtual synchronous generator technology is introduced into the inverter control system. Unlike the synchronous generator, its parameters are flexible and adjustable, which can enhance the robustness of the microgrid. This paper firstly analyzes the influence of the moment of inertia and damping on the system, it is based on the equivalent relationship between the three-phase inverter and the synchronous machine, combined with the VSG rotor motion equation, and conducts stability analysis based on the small signal model and root locus. Based on the existing control strategy, a virtual synchronous generator cooperative control strategy based on the optimal damping ratio is proposed. Finally, the VSG collaborative control model is built in MATLAB/Simlink, the experimental results show that the proposed control strategy effectively improves the transient performance of the VSG, the system active power overshoot is reduced from 5% to 0, and the frequency deviation peak is lowered from 0.174 Hz to 0.115 Hz, which confirms the effectiveness of the proposed control strategy.

Rapid advances of sensing and cloud technologies transform the manufacturing system into a data-rich environment and make production scheduling increasingly complex. Traditional offline scheduling methods are limited in the ability to handle low-volume-high-mix workorders with diverse design specifications. Simulation-based methods show the promise for distributed scheduling of manufacturing jobs but are mostly implemented with historical data and empirical rules in a static manner. Recently, artificial intelligence (AI) algorithms fuel increasing interests to solve dynamic scheduling problems in the manufacturing setting. However, it's difficult to utilize high-dimensional data for production scheduling while considering multiple practical objectives for smart manufacturing (e.g., minimize the makespan, reduce production costs, balance workloads). Therefore, this paper presents a new AI scheduler with composite reward functions for data-driven dynamic scheduling of manufacturing jobs under uncertainty in a smart factory. Internet-enabled sensor networks are deployed in the smart factory to track real-time statuses of workorders, machines, and material handling systems. A novel manufacturing value network is developed to take high-dimensional data as the input and then learn the state-action values for real-time decision making. Based on reinforcement learning (RL), composite rewards help the AI scheduler learn efficiently to achieve multiple objectives for production scheduling in real time. The proposed methodology is evaluated and validated with experimental studies in a smart manufacturing setting. Experimental results show that the new AI scheduler not only improves the multi-objective performance metrics in the production scheduling problem but also effectively copes with unexpected events (e.g., urgent workorders, machine failures) in manufacturing systems.

In this paper, an improved online particle swarm optimization (PSO) is proposed to optimize the traditional search controller for improving the operating efficiency of the permanent magnet synchronous motor (PMSM). This algorithm combines the advantages of the attraction and repulsion PSO and the distributed PSO that can help the search controller to find the optimal d - axis air gap current quickly and accurately under non-stationary operating conditions, thereby minimizing the air gap flux and then improving the motor efficiency. To verify the effectiveness and stability of this proposed algorithm, the operating efficiency of PMSM as using this proposed algorithm is compared with that of traditional search controller under non-stationary operating conditions. The results show that the proposed algorithm can improve the operating efficiency of PMSM by 6.03% on average under non-stationary operation conditions. This indicates that the search controller based on the improved PSO has a better adaptation to the variation of external operating conditions, and can improve the operation efficiency of PMSM under non-stationary condition.

Power distribution centres distribute power to motor control centres in industrial facilities, as well as supply power to medium voltage drives among other loads. In reliability studies, the use of the combinatorial methods to evaluate the reliability of IEC-61850 based SCN in power distribution centres is preferred, owing to the ease of comprehension and applying the methods. The disadvantage of the methods is that repairs are assumed to be perfect and that all system faults are known; whereas it is not the case in practice. Markov model enables the impact of imperfect repairs on the system reliability performance to be investigated using the eigenvalue analysis method based on the concept of linear dynamical systems. The method presented in this paper advances the eigenvalue analysis method and focusses on the incremental responsiveness of the system to imperfect repair factors based on absorbing Markov chain and matrix calculus. The results indicate that even though the system is dependent on repair factors, the system is perfectly inelastic to the repair efficiency factors. However, the diagnostic coverage of the system is the most critical of the two factors, with higher elasticity as the factor approaches 100%. The results also indicate that both sensitivity and elasticity rapidly decrease as the diagnostic coverage of the system decrease. Thus, it is concluded that more emphasis must be put on the system diagnostic coverage because it is embedded in the system design, which can be expensive and practically not achievable once the system is commissioned and is running.

Existing night vehicle detection methods mainly detect vehicles by detecting headlights or taillights. However, these features are adversely affected by the complex road lighting environment. In this paper, a cascade detection network framework FteGanOd is proposed with a feature translate-enhancement (FTE) module and the object detection (OD) module. First, the FTE module is built based on CycleGAN and multi-scale feature fusion is proposed to enhance the detection of vehicle features at night. The features of night and day are combined by fusing different convolutional layers to produce enhanced feature (EF) maps. Second, the OD module, based on the existing object detection network, is improved by cascading with the FTE module to detect vehicles on the EF maps. The proposed FteGanOd method recognizes vehicles at night with greater accuracy by improving the contrast between vehicles and the background and by suppressing interference from ambient light. The proposed FteGanOd is validated on the Berkeley Deep Drive (BDD) dataset and our private dataset. The experimental results show that our proposed method can effectively enhance vehicle features and improve the accuracy of vehicle detection at night.

Sketch face recognition refers to the process of matching sketches to photos. Recently, there has been a growing interest in using deep learning to learn discriminative features for sketch face recognition. However, the success of deep learning relies on the large-scale paired images to counteract effects such as over-fitting, since the amount of the paired training data is relatively small, the discriminative power of the deeply learned features will inevitably be reduced. This paper proposes a novel deep metric learning method termed domain alignment embedding network for sketch face recognition. Specifically, a training episode strategy is designed to alleviate the small sample problem, and a domain alignment embedding loss is proposed to guide the feature embedding network to learn discriminative features. Extensive experimental results on the UoM-SGFSv2 and PRIP-VSGC datasets are verified to show the effectiveness of the proposed method.

Convolutional Neural Networks (CNNs) have been shown to be very useful in image recognition and other Artificial Intelligence (AI) applications, however, at the expense of intensive computation requirement. To address the challenge of overwhelming calculation requirements, researchers have proposed various network pruning techniques. But, due to the irregular sparse patterns, unstructured sparse networks are difficult to compute efficiently on either Graphic processing units (GPUs) or Field Programmable Gate Arrays (FPGAs). In this paper, we propose a software/hardware co-optimized Reconfigurable Sparse convolutional Neural Network accelerator design (RSNN) on FPGAs. A novel sparse convolution dataflow is proposed with simpler control logic than existing mux-based selection logic. To balance the computation load on different Processing Units (PUs), we propose a software-based load-balance aware pruning technique as well as a kernel merging method. Experimental results show that RSNN is $2.41\times - 7.91\times $ better on Digital Signal Processor (DSP) efficiency than previous dense CNN FPGA accelerators, and $1.23\times - 2.93\times $ better than previous sparse CNN FPGA accelerators.

High-performance DC interfaces play a pivotal role in microgrids with distributed energy resources (DERs), which would significantly improve the utilization of DERs and the flexibility of the prosumers. For a candidate DC interface converter, how to maintain the performance over a wide operation range becomes the biggest challenge. This paper proposes a flexible structure soft-switching three-level (TL) dc-dc converter, which adds four high-speed MOSFETs and one low-speed electronic relay to achieve overall optimum performance over a wide range. Under high output or low input voltage conditions, the relay and the extra MOSFETs on the primary side are ON permanently. The proposed converter is a zero-voltage switching (ZVS) converter, which guarantees ZVS for all primary switches even under 0 load currents. Under low output or high input voltage conditions, the relay and the extra MOSFETs in the secondary side are OFF permanently. The proposed converter is a zero-voltage and zero-current switching (ZVZCS) converter, and two added MOSFETs provide ZVZCS operation not only for the main primary switches but also themselves over a wide range. Other advantages of the presented converter include reduced filter size, no primary side circulating currents, and reduced current stress of the clamping capacitor. Besides, the switching loss caused by the added power devices is low because of the full range of soft-switching operation and low on-state resistance of these devices. This paper discusses the circuit configuration, operation principle, soft switching characteristics, technical comparison, and experimental results from a 1-kW prototype prove the rightness of the presented converter.

Saturated power demand is a crucial indicator affecting the scale of power grid development. It is essential for power grid planning to study the saturated trend of power demand and predict the time and scale when the order reaches saturation. In this context, this paper divides the growth process of power demand into three stages, proposes a modified self-adaptive Logistic model, and takes the power demand in East China as an example to verify the accuracy and practicability of the model, and carries out the saturation prediction research. Finally, according to the research results, the corresponding suggestions are put forward.

This paper presents an adaptive backstepping sliding mode tracking control method for underactuated unmanned surface vehicle. The tracking controller is designed by combining backstepping and adaptive sliding mode control technology. The virtual control law is used to replace the control strategy of position error, and the position tracking control is transformed into velocity control, which can effectively avoid the input saturation and singular value problem existing in the design of traditional backstepping control law. The neural shunt model method is used to solve the differential explosion problem caused by virtual control law. The adaptive technology based on backstepping sliding mode theory is introduced to compensate the model uncertainty and time-varying disturbances of the system, and the robustness of the underactuated USV is enhanced by this method in the unknown environment. Based on the Lyapunov stability theory, it is proved that the control system error is ultimately uniformly bounded. The simulation results show that the proposed adaptive backstepping sliding mode tracking control method can achieve stable tracking in the case of system parameter uncertainty and time-varying disturbances, so it can be concluded that this method is reasonable and effective.

The rapid development of renewable energy and the continuous growth of peak load bring new challenges to the dispatching capacity of generation side. In view of the possible mismatch between power generation of renewable energy and the load, we propose an integrated optimal dispatching strategy model of power generation and consumption interaction in this paper. Among them, the generation side resources include wind power, photovoltaic and battery energy storage and the load side dispatching resources include transferable load, interruptible load and electric vehicles. The model takes the benefits of generation side and the benefits of load side as objectives. Moreover, three different dispatching orders are put forward, which are random dispatching, interruption and transfer of load before electric vehicles charging and discharging, electric vehicles charging and discharging before interruption and transfer of load. In addition, we propose seven different dispatching combinations based on the optimal dispatching order to analyse the response ability of different load side dispatching resources. Multi-objective Genetic Algorithm and the method of technique for order preference by similarity to an ideal solution are applied in this study to obtain the optimal results. It can be identified that the response ability of different load side dispatching resources is different, and the overall dispatching can improve the benefits of generation side and the benefits of load side. Meanwhile, the proposed strategy considers the willingness of the load to the greatest extent and effectively improves the utilization rate of wind and solar power generation.

With the rapid development of mobile internet technology, mobile applications (apps) have been rapidly popularized. To facilitate users' choice of apps, app recommendation is becoming a research hotspot in academia and industry. Although traditional app recommendation approaches have achieved certain results, these methods only mechanically consider the user's current context information, ignoring the impact of the user's previous related context on the user's current selection of apps. We believe this has hindered the further improvement of the recommendation effect. Based on this fact, this paper proposes a novel context-aware mobile application recommendation approach based on user behavior trajectories. We named this approach CMARA, which is the initials acronym of the proposed approach. Specifically, 1) CMARA integrates the heterogeneous information of the target users such as the user's app, time, and location, into users behavior trajectories to model the users' app usage preferences; 2) CMARA constructs the context Voronoi diagram using the users' contextual point and leverages the context Voronoi diagram to build a novel user similarity model; 3) CMARA uses the target user's current contextual information to generate an app recommendation list that meets the user's preferences. Through experiments on large-scale real-world data, we verified the effectiveness of CMARA.

This paper is concerned with the problem of event-triggered non-fragile H∞ filter design for interval type-2 (IT2) fuzzy systems subject to output quantization. The nonlinear plant is efficiently described by an IT2 fuzzy model, and the lower and upper membership functions with weighting coefficients are employed to characterize parameter uncertainties in the plant. To enhance filter adaptability, an IT2 non-fragile fuzzy filter method is proposed to acquire available information for the unknown state of the plant, where the multiplicative gain variations is taken into account in the filter analysis design process. In order to avoid continuous communications and save limited bandwidth, a dynamic event-based mechanism is employed to adopt the limited communications links. Then, based on the Lyapunov theory together with the inequality technique, a filtering system with event-based mechanism and measurement quantization is analyzed and constructed. Moreover, the obtained sufficient conditions for system analysis are given in the form of linear matrix inequalities (LMIs). Finally, a numerical simulation is provided to verify the effectiveness of the proposed filter design strategy.

Due to the potential for significant process benefit, microwave assisted concrete recycling is highlighted by many scholars and engineers over the world recently. In this paper, a multi-field model is established to simulate the microwave heating process of concrete specimen with single-particle aggregate by using of COMSOL Multiphysics. The validation of this model is verified by comparison with temperature from experiment. Mechanical properties of aggregate-mortar interface during microwave heating process was experimentally investigated. Electric, temperature, stress field evolution and moisture transformation during microwave heating process are discussed in detail. Temperature and stress gradient at mortar-aggregate interface influenced by microwave heating power, frequency and duration are analyzed and graphically presented, which are the dominant factors to achieve the interface debonding. Based on the comparison between concrete absorbed energy and temperature variation, energy efficiency for microwave heating concrete is also briefly discussed. Results of this investigation could provide a multi-field understanding for the microwave treatment on concrete.

In order to ensure the reliability and stability of the manufacturing process, tool wear state should be realized real-time and accurate monitoring. This paper proposes a tool wear state recognize and predictive framework model based on Stacking Sparse De-noising Auto-Encoder (SSDAE), the Particle Swarm Optimization (PSO) and the Least Squares Support Vector Machine (LSSVM). The Stacking Sparse De-noising Auto-Encoder (SSDAE) technique is utilized to realize multi-feature signal dimension reduction with the aim of improving the prediction accuracy, which reduces the dependence on the prior knowledge of feature selection and greatly improves the modeling efficiency. PSO technique is helpful for adaptive optimization of kernel parameters, which greatly improved computing power and LSSVM model prediction accuracy. A dataset from a real machining process is utilized to verify the effectiveness of proposed model in improving the prediction accuracy. The experimental results show that a high correlation coefficient greater than 0.95 is used to extract feature vector from time domain, frequency domain and time-frequency domain three directions, and the proposed SSDAE-PSO-LSSVM model performs better than Partial Least Squares Regression (PLSR), Back Propagation Neural Network (BPNN) and Extreme Learning Machine (ELM) in terms of prediction accuracy.

Complex systems can be characterized by their level of order or disorder. An ordered system is related to the presence of system properties that are correlated with each other. For example, it has been found in crisis periods that the financial systems tend to be synchronized, and symmetry appears in financial assets' behavior. In retail, the collective purchasing behavior tends to be highly disorderly, with a diversity of correlation patterns appearing between the available market supply. In those cases, it is essential to understand the hierarchical structures underlying these systems. For the latter, community detection techniques have been developed to find similar behavior clusters according to some similarity measure. However, these techniques do not consider the inherent interactions between the multitude of system elements. This paper proposes and tests an approach that incorporates a hierarchical grouping process capable of dealing with complete weighted networks. Experiments show that the proposal is superior in terms of the ability to find minimal energy clusters. These minimum energy clusters are equivalent to system states (market baskets) with a higher probability of occurrence; therefore, they are interesting for marketing and promotion activities in retail environments.

A new robust cubature Kalman filter is proposed using adaptive generalized maximum correntropy (AGMC) criterion rather than the conventional MMSE criterion in this paper. In the proposed method, the adaptive generalized maximum correntropy (AGMC) criterion is firstly constructed from an adaptive forgetting correntropy based cost function, which is rather robust with respect to the process uncertainty and non-Gaussian noise. On this basis, a new robust cubature Kalman filter is further derived, where the predicted state vector and received measurements are processed simultaneously based on the regression form derived via the statistical linearization approach. An adaptive forgetting scheme is then proposed in combination with the AGMC-CKF to update the parameters of the AGMC adaptively in real time. Taking advantage of the AGMC, the unknown noise statistics caused by the process uncertainty and non-Gaussian noise can be effectively suppressed. Simulations and car-mounted experiments demonstrate that the proposed filter is superior in terms of estimation accuracy and robustness as compared with the related state-of-art methods.

Energy harvesting (EH) and spectrum harvesting (SH) are two promising and useful green communication and networking mechanisms for the next-generation wireless networks. While the former techniques exploit ambient energy sources to scavenge energy, the latter exploit the unused or moderately used electromagnetic spectrum. With the advent of cyber-physical systems and the Internet-of-Things (IoT), the presence of tens of billions of low power sensor devices would soon be a reality. These small sensing devices would be present in many systems around us, such as home appliances, telecommunication devices, medical electronics, transport systems, etc. These miniaturized, low-power consuming devices may exploit EH and SH techniques for energy storage and communication. These EH-SH-enabled sensors or low-power nodes need to consume very little energy for sensing and communicating opportunistically. However, several theoretical problems and practical challenges exist in EH-SH communications. In this comprehensive survey paper, we first present the historical background of EH, and SH techniques, and their development over several decades. Specifically, we focus on EH-SH communication technologies and protocols for a wide range of systems and networks. We present a detailed survey of the various harvesting techniques and protocols from recent literature. Finally, we describe exciting open, intra-disciplinary, and inter-disciplinary challenges for further research on EH-SH communication technologies.

Under the framework of event triggering, this paper discusses a novel integral sliding mode (ISM) anti-disturbance algorithm for a class of nonlinear systems with mismatched disturbances. For the sake of describing more irregular disturbances, T-S fuzzy disturbance models are imported by choosing different basis functions and the corresponding disturbance observer (DO) is then put forward to realize the estimation of unknown disturbances. By utilizing the trigger threshold condition with the estimation information, both the integral sliding surface (ISS) and the event-triggered ISM controller are designed for compensating mismatched disturbances. It is guaranteed that the state trajectories can reach the ISS within a pre-designed finite time. Moreover, not only the system stability, but also the favorable tracking performance and the boundedness of the minimum trigger interval are simultaneously be achieved. Finally, the simulation results of an A4D aircraft model further embody the effective of designed control algorithm.

In this paper, the axis-aligned single mode- four mode- single mode (SFS) fiber structure filter is proposed to be an all-optical temporal differentiator. It's shown that differentiators with different orders can be achieved by adjusting the parameters of the axis-aligned SFS fiber structure filter. All-optical temporal differentiators with orders of 0.375, 0.77 and 1 are obtained by using the axis-aligned SFS fiber structure filter, of which processing errors are 8.08%, 2.81%, 0.74%, respectively. The proposed design provides an easy and competitive way for the realization of the all-optical temporal differentiator.

Among various biometric methods, palm vein authentication has taken significant attention because of its uniqueness, stability and non-intrusiveness. In this paper, we propose a palm vein authentication model using convolutional neural networks (CNN), which is the most popular deep learning architecture and Bayesian optimization. First and foremost, region of interest (ROI) of the palm vein is extracted as an image and filtered by Jerman enhancement filter to enhance the gray levels of the vein patterns. The proposed CNN model allows different numbers of convolutional layers to be added to optimize the network structure. Furthermore, the model is trained with training data to extract the highly representative features of the different classes. The training process is performed at every objective function evaluation, each with a different network structure and training options using a Bayesian optimization algorithm to find the optimal network structure and training options in a search space of possible solutions. The CNN model serves as the palm vein template creator or feature extractor for our identification and verification experiments. Receiver operating characteristic (ROC) curve and equal error rate (EER) were plotted for evaluating the performance of the proposed model. Our proposed method attained an average identification accuracy of 99.4 % and average EER of 0.0683%, which outperforms state-of-the-art palm vein authentication approaches.