Search

The limitations of scaling in CMOS technology pose challenges in meeting the requirements of future applications. To address these challenges, researchers are exploring various design techniques, including Approximate Computing (AC), which leverages the inherent error resilience of applications to achieve high performance and energy gains with desired quality. AC has gained popularity as a computer paradigm for error-resilient applications, and many researchers have studied AC across computing layers and developed tools for implementing these techniques. This paper provides a comprehensive survey of AC techniques at the abstraction levels of software and hardware and discusses the tools to implement AC in hardware and software, quality evaluation tools and comparison points. The paper also covers existing frameworks for AC, potential applications, future research directions, challenges and limitations. This information can guide researchers in identifying promising avenues for further advancements and innovations in this domain. Additionally, this paper compares state-of-the-art surveys of AC and highlights the unique features and contributions of this work that distinguish our work from previous surveys. [ABSTRACT FROM AUTHOR]

Recent advances in speaker embeddings promote a great development of speaker diarization. However, determining 'who spoke when' in the meeting scenarios is still challenging due to similar speaker voices and unknown speaker quantity. In this paper, this research proposes enhanced discriminative features for speaker diarization, including discriminative speaker-specific features based on Siamese networks, and a speaker re-verification method. With Siamese architecture, SiamTDNN, this research first explores latent representations which is capable of modeling intra-class and inter-class differences between speakers, by training with audio pairs. Then, the re-verification method is introduced with a local-global strategy to identify speakers in a multi-person talking scene. Our method provides a novel speaker embedding with enhanced discriminative power for disambiguated speakers and achieves an elevated upper bound on the number of speakers. The proposed speaker embedding achieved an EER of 1.1% and a minDCF of 0.1192 on VoxCeleb1 for the speaker verification task. Extensive experiments on AiShell-4, ICSI, AMI and VoxConverse demonstrate the effectiveness of the proposed method with an average DER reduction of 3% and an RTF of 0.0792. [ABSTRACT FROM AUTHOR]

This paper describes a four-port MIMO antenna array design featuring bow-tie-shaped slot-loaded patches with wideband capabilities that cover the frequency range from 24.2 GHz to 30.8 GHz. The proposed antenna design is printed on an FR4 substrate and occupies an area of 25 × 24 mm2. The MIMO antenna consists of four antenna arrays that are symmetrically placed in an upper-lower configuration. The bow-tie-shaped slots loaded radiators are separated horizontally by 3.48 mm and vertically by 5.94 mm. Each antenna array contains two elements that are separated by a distance of wavelength/4. The suggested MIMO antenna array delivers a high gain of 19.09 dB at 27.8 GHz and has a bandwidth of 6.6 GHz that covers the frequency band of 24.2–30.8 GHz. The research demonstrates the quality of the proposed MIMO antenna through various diversity parameters such as mutual coupling, port correlation, diversity gain, and data rate that can be transmitted over a communication medium. The simulation results are validated and found to be consistent with the experimental results. The presented antenna covers the entire bandwidth allocated to different regions, including Europe (24.25–27.5 GHz), Sweden (26.5–27.5 GHz), USA (27.5–28.35 GHz), China (24.25–27.5 GHz), Japan (27.5–28.28 GHz), and Korea (26.5–29.5 GHz). The proposed MIMO antenna design could be an excellent option for 26/28 GHz 5G NR n257, n258, and n260 bands under mm-wave wireless communication systems. [ABSTRACT FROM AUTHOR]

As the basic and core component of electronic systems, CPU security is extremely important to network security. Even an unremarkable faulty instruction on the CPU may lead to serious security problems, such as the operating system crashes or privilege increase since it is often considered as a trusted black box. Therefore, CPU instruction detection is particularly crucial to CPU security. However, most existing methods of CPU instruction detection, based on the inconsistency of microarchitecture and instruction set design, suffer from slow speed and low accuracy. Our work is motivated to propose a practical framework for searching CPU undocumented instruction with fast speed and high accuracy. In this paper, we put forward a general framework InsFinder to detect undocumented instruction on CISC and RISC CPU by an efficient and accurate fuzzing method. It makes use of the instruction format to make advanced predictions, which greatly reduces the search space. Moreover, by introducing classification, de-redundancy, and verification, InsFinder greatly improves the detection accuracy. Experiments show that compared with the existing methods, InsFinder is more effective which costs at least 50% less processing time in detecting undocumented instructions on x86-64, ARM64, and RISC-V, and more accurate which divided the detection results into 4 categories. After filtering, the detection results were reduced from millions to less than 10,000. [ABSTRACT FROM AUTHOR]

First-order allpass filters are analog filters with a unit magnitude response, but they change the phase shift between input and output signals at various frequencies. This paper presents the transadmittance-mode first-order allpass filters based on the modified current-controlled current differencing transconductance amplifier (M-CCCDTA). The proposed filters utilize a single M-CCCDTA and a grounded capacitor with no external resistors, making them well suited for implementation in an integrated circuit. The gain and phase response of the proposed filters can be adjusted electronically and separately. Also, the high output impedance of the proposed filters at both the input voltage node and the output current node makes cascading easy without the need for buffer devices. A quadrature sinusoidal oscillator based on the proposed first-order allpass filter has been designed as an example of an application. The PSpice simulation and actual experiments are utilized to validate the functionality of the proposed filters. The simulation and experimental results are consistent with the idea of anticipation. [ABSTRACT FROM AUTHOR]

A new universal active filter configuration is presented in this paper which is capable of realizing all three first-order filter responses namely all pass (AP), low pass (LP) and high pass (HP) using a single second generation voltage conveyor (VCII+), a virtually grounded capacitor and two resistors. The proposed circuit offers tunability of the cut-off frequency, gain (for HP filter) and simultaneous availability of two filter outputs, one at a low impedance node (HP/AP), the other output, namely the LPF will require a voltage buffer to avoid loading. The nonideal analysis of filter responses based on the nonideal model of VCII+ is carried out by considering all the parasitic immittances and nonideal gains and found that they have no significant effect on the performance of the proposed filter structure. Sensitivity analysis with respect to active and passive components has also been carried out. The functionality of the presented circuit has been validated through simulations and experimental results, where the VCII+ was implemented using the macro model of the commercially available CFOA IC AD844. Additionally, the proposed circuit has been tested using a CMOS VCII+ implemented with 0.18 μ m TSMC technology parameters. The tests include advanced evaluations based on Monte Carlo simulations. [ABSTRACT FROM AUTHOR]

Doppler frequency carries the information about the relative velocity of the object regarding the radar antenna. However, since the target maneuvers can be temporary with high intensity, Doppler frequency has to be estimated by using a window function in the frame of time-dependent Fourier transform. In order to minimize the estimation error, the window function width should be adaptive. The window length adaptation has been performed based on the estimates of target acceleration and signal-to-noise ratio (SNR). In this paper, we give focus on the SNR estimation and propose a new approach based on autoregressive method of spectral estimation, where in one step the amplitude of the sinusoidal signal and noise variance are estimated. The simulation results justify the advantage of the proposed approach in general applications of signal processing of one sinusoid in white noise environment. [ABSTRACT FROM AUTHOR]

As a special type of location-based service (LBS), crowdsensing becomes more prosperous in people's daily life. However, during the process of task distribution, the publisher's and workers' locations will be revealed to each other, and then their personal privacy is violated. So in this paper, in order to cope with the violation of location privacy in crowdsensing and provide privacy preservation service for both entities, an active oblivious transfer-based location privacy preservation crowdsensing scheme (short for AOTC) has been proposed. In this scheme, the oblivious transfer is used to encrypt the range of sensing grid of workers, and then matching sensing grids with the sensing region of the publisher without decryption. During the whole process, the process of location matching and results sending is disposed of by the entity of workers actively, so does not establish any data aggregation that can be used as the point of attack. As a result, the AOTC can guarantee the personal privacy of both entities in crowdsensing cannot be obtained by each other, and guarantee other workers also difficult to obtain the precise location of any workers. In addition, as workers send the sensing result to the publisher actively this scheme can also increase the probability of workers' participation potentially. At last, the theoretical privacy preservation ability of AOTC is analyzed in the section on security analysis with three types of privacy threats. Then the performance of AOTC is compared with other similar schemes in both privacy preservation and execution efficiency, so in simulation experiments, comparison results with brief analyses will confirm that the AOTC has achieved the desired effect and will further demonstrate the superiority. [ABSTRACT FROM AUTHOR]

Feature extraction and matching are the basic procedures of the so-called Visual Odometer (VO), Simultaneous Localization and Mapping (SLAM) and many other image processing algorithms. Oriented features from accelerated segment test (FAST) and rotated binary robust independent elementary features (ORB) algorithm are widely used since they are computationally faster. In this paper, we proposed a method to generate a value for a feature, the value is called signature. In the matching step, we only compute Hamming distances of descriptors with the same signatures. Hence, the matching time is shortened. Compared with the original ORB algorithm, features to be matched dropped 69.63% on TUM datasets and 85.7% on VGG datasets by adopting our strategy. In addition, the precision is above 85% on both VGG and TUM datasets. We design a customized hardware architecture for ORB feature extraction and matching based on the proposed method. The hardware structure is implemented on Xilinx ZCU102 evaluation board. The clock frequency is set to 150 MHz. Our Field Programmable Gate Arrays (FPGA) system achieves 193 fps on 1 2 8 0 × 7 2 0 images with 1984 features on average and 314 fps on 6 4 0 × 4 8 0 images with 700 features on average, which is more efficient compared to the state-of-the-art works. [ABSTRACT FROM AUTHOR]

An adaptive biased architecture of voltage differencing transconductance amplifier (AB-VDTA) with high transconductance gain is proposed in this paper. The proposed AB-VDTA is very efficient in terms of power. The structure proposed involves of two units namely two transconductance amplifiers (TAs) and two squarer. The bias current of the TA is made to vary with a square relation of the differential input voltage. Therefore, the proposed structure provides tunable gain depending on the input differential voltage. The proposed AB-VDTA exhibits improved transconductance gain, transient characteristics, reduced standby power dissipation and linearity for large range of inputs. The mathematical formulation has been presented to establish the characteristics of the proposed AB-VDTA. The proposed structure of AB-VDTA is validated through SPICE simulations using 180 nm complementary metal oxide semiconductor (CMOS) technology. The proposed scheme has an edge over the existing ones as the outlined methods enhance the transconductance gain by increasing the bias current. The g m 1 values are observed to be 2.3 and 1.4 mS for proposed AB-VDTA and conventional VDTA, respectively, with the corresponding 3 dB frequencies 620 and 348 MHz. Therefore, 64% improvement in transconductance gain is recorded for same value of bias currents. The linear input range of TA1 is observed to be ± 1 5 0 mV and the overall linear range of the VDTA is ± 1 0 0 mV for the proposed AB-VDTA. The PVT analysis is carried out to show the effect of process corners. To check the robustness of the proposed VDTA, Monte Carlo analysis is performed, and results have been included in the form of histograms. As an application example, a current mode (CM) universal single input multiple output (SIMO) biquad filter is also designed using the proposed VDTA to show its usefulness, and a 2.7 times higher pole frequency is obtained at equal bias current. [ABSTRACT FROM AUTHOR]

This paper proposes a point-of-interest (POI) sequence recommendation algorithm based on BERT-ACNN-GRU to address the issues faced by the existing POI recommendation model in social network large data, such as the difficulty in extracting deep feature information and the low recommendation performance. Firstly, the semantic relationship between a word and its context in the text is combined using the bidirectional encoder representation from transformers (BERT) model to effectively eliminate the influence of word distance and obtain the contextualized word vector. Secondly, a convolutional neural network (CNN) utilizing a gated recurrent unit (GRU) is employed to capture the feature information of the text. Lastly, the attention method is utilized to assign weight scores to various terms in order to provide more attention to particular words and boost the precision of recommendations. The experiments demonstrate that the precision, recall rate, F1 score and mean average precision (mAP) of the proposed method are 0.097, 0.26, 0.103, and 0.085 on the Gowalla dataset when the recommendation list has a length of 10, respectively. On the Yelp dataset, the precision is 0.093, the recall rate is 0.26, F1 is 0.099, and MAP is 0.089. Hence, the proposed method can effectively enhance the performance of the POI recommendation system. [ABSTRACT FROM AUTHOR]

In recent years, cloud computing technologies have been developed rapidly in this computing world to provide suitable on-demand network access all over the world. A cloud service provider offers numerous types of cloud services to the user. But the most significant issue is how to attain optimal virtual machine (VM) allocation for the user and design an efficient big data storage platform thereby satisfying the requirement of both the cloud service provider and the user. Therefore, this paper presents two novel strategies for optimizing VM resource allocation and cloud storage. An optimized cloud cluster storage service is introduced in this paper using a binarization based on modified fuzzy c-means clustering (BMFCM) algorithm to overcome the negative issues caused by the repetitive nature of the big data traffic. The BMFCM algorithm utilized can be implemented transparently and can also address problems associated with massive data storage. The VM selection is optimized in the proposed work using a hybrid COOT-reverse cognitive fruit fly (RCFF) optimization algorithm. The main aim of this algorithm is to improve the massive big data traffic and storage locality. The CPU utilization, VM power, memory dimension and network bandwidth are taken as the fitness function of the hybrid COOT-RCFF algorithm. When implemented in CloudSim and Hadoop, the proposed methodology offers improvements in terms of completion time, overall energy consumption, makespan, user provider satisfaction and load ratio. The results show that the proposed methodology improves the execution time and data retrieval efficiency by up to 32% and 6.3% more than the existing techniques. [ABSTRACT FROM AUTHOR]

In this paper, the six degrees of freedom adaptive tracking control for chaser spacecraft with uncertainty and disturbance to approach a noncooperative target are studied. The system equations with coupled relative translation and relative rotation are modeled. Considering the unknown inertia uncertainties and state-dependant disturbances, a backstepping-based adaptive tracking control law is developed to achieve high-precision proximity operations between chaser and noncooperative target. The unknown inertia and disturbance are estimated by adaptive update laws. The properties of the developed control law are discussed through the comparison of simulation results. [ABSTRACT FROM AUTHOR]

Urban user water demand prediction (WDP) is of significant importance for smart water supply system, which can provide a strong decision-making basis for the dispatching and management of smart water supply system. However, owing to the fluctuation, intermittence and nonstationarity of the user's water consumption in urban buildings, it is extremely difficult to predict accurately. Therefore, a novel short-term WDP model (Singular Spectrum Analysis Convolutional Neural Network Bidirectional Gate Recurrent Unit, SSA-CNN-BiGRU) is proposed to promote the stability and accuracy of WDP, which successfully introduces organic combinations including deep learning, decomposition technique, and data partitioning policies into the domain of WDP. First, raw data are decomposed into components that carry distinct frequency signals for weakening its nonstationarity and complexity. Then, all the components are automatically divided into several groups using clustering algorithm based on their entropy, after which deep learning method is adopted to predict by groups. Finally, the predicted result of each group is summed up to be fused as the final value. To validate the predictive performance of SSA-CNN-BiGRU, real data have been selected for this study. In experiments, SSA-CNN-BiGRU achieved a fitting of 94.73%. Comparison by relevant evaluation metrics demonstrates that the proposed model exhibits superior performance, thus providing a more accurate basis for WDP. [ABSTRACT FROM AUTHOR]

This paper presents a Schmitt trigger (ST) circuit with anti-clockwise hysteresis and its utility as an adjustable square/triangular wave generator. The ST circuit employs a single dual-X second-generation current conveyor and two resistors. The threshold voltages of the circuit are adjustable such that it can also function as a zero crossing detector. The circuit has a wide operational frequency range and also consumes low power. Without using an additional active block, a square/triangular wave generator circuit is also realized within the same circuit topology. The generator circuit uses a single active element, one grounded resistor and one grounded capacitor only. The use of grounded passive components makes the proposed generator circuit easily integrable. Additionally, the generator circuit is adjustable as its duty cycle is tunable by means of an external current source. The generator circuit also has a wide operational frequency range from 2.1 Hz to 19.2 MHz. The theoretical aspects of the proposed circuits are validated via Cadence simulations. Additionally, a prototype of DXCCII, which is implemented by using current feedback operational amplifier ICs AD844, is used to verify the ST experimentally. [ABSTRACT FROM AUTHOR]

In this paper, the combination of a high step-up boost converter with extended voltage multiplier cell and isolated SEPIC converter with a series output module is presented. Since these converters use the same current source in their inputs, they can share the common parts, so the structure of the proposed converter is simple. In this converter, the high frequency transformer used in the SEPIC converter and the inductor of the extended voltage multiplier cell are coupled, so the volume of the magnetic parts is reduced. In addition, an active-clamp circuit is used to solve the problem of leakage inductor and achieve to Zero Voltage Switching (ZVS) turn on/off for switches and ZCS for diodes. Compared to its high step-up counterpart, the proposed converter has the following: (1) distributed voltage stress on the diodes and voltage of the output capacitors are fair. (2) Higher voltage gain. (3) Voltage stresses on the diodes are lower and the reverse-recovery problem of the diodes is eliminated. (4) Minimum root mean square (RMS) leakage inductor current. The operational principle and characteristics of the proposed converter are presented, and in order to verify the proposed converter, a 200 W, 20–400 V prototype converter is implemented and experimental results are provided. [ABSTRACT FROM AUTHOR]

Nowadays, multilevel inverter (MLI)-based induction heating (IH) technology is the best-suited option in many domestic, commercial and medicinal applications because of its higher efficiency, low cost, fast heating and lower electromagnetic interference (EMI). This paper presents reduced components count high-frequency (HF) single phase five-level (5-L) MLI for IH applications. The working methodology of the proposed system has been analyzed in the multicarrier phase disposition pulse width modulation (MC-PD-PWM) method in high switching frequency i.e., 10 kHz. In addition, the proposed topology can produce output voltage twice of input voltage without any external source. Moreover, the detailed power loss and reliability analysis have been evaluated for the proposed system. In terms of switch count and efficiency, the single-phase 5-L inverter architecture is compared to some conventional and current topologies. Finally, to validate and justify the operation of the proposed system, the simulation and experimental results of five-level inverter are presented. [ABSTRACT FROM AUTHOR]

In this research on range query technology for two-tiered wireless sensor networks, the existing researches still need improvement in privacy security and communication cost of the sensor node. In this paper, a security multi-dimensional range query protocol is proposed. The left 0-1 encoding mechanism (a new and optimized encoding mechanism), HMAC algorithm, and BASE64 encoding are employed to optimize the calculation of comparison factor. The encryption index chain is constructed by using AES algorithm to replace the long encryption constraint chain, so as to lower the energy consumption of computation and communication. The base station can verify the authenticity and integrity of query results on the basis of continuity of storage time and encryption index chain. The experiment has been conducted in the real systems. The protocol is compared and analyzed with the latest proposed Communication-Efficient Secure Range Queries (CSRQ). The results have shown that the proposed protocol can gain a better performance in terms of four aspects, i.e., the length of sensing data, the number of data collected in each cycle, the number of data dimensions and the number of sensor nodes in each unit. Meanwhile, compared with CSRQ, the communication cost of the proposed protocol is reduced by at least 10%. [ABSTRACT FROM AUTHOR]

In medical image examination, image segmentation is the broadly used method. Currently, the efficient segmentation of mammogram images is the main challenge. Many methods were presented for segmenting the mammogram images, but the results are not satisfactory. In this paper, an efficient segmentation of mammogram images-based Multilevel Thresholding (MLT) method is proposed. Initially, the preprocessing step is executed for eliminating the unnecessary noises. For gaining the useful features from the mammogram images, mammogram image segmentation is carried out using multilevel thresholding method. In this paper, a novel Multi-Objective Emperor Penguin Optimization (MOEPO) algorithm is proposed for searching the multilevel greatest thresholds that segment the images into background and objects. The objective functions of the MLT are Otsu's method, Kapur and Tsallis entropy. The effectiveness of the proposed method is analyzed using several performances evaluating metrics, like PSNR, FSIM and SSIM. The experimental outcomes show that the performance of the proposed technique is superior to other state-of-the-art methods. The proposed technique is likened to three existing models, viz. ScPSO-MT, Double Threshold and IWO-SUSAN. The SSIM of the proposed technique is 24.99%, 27.83% and 26.95% better than ScPSO-MT, Double Threshold and IWO-SUSAN existing approaches. The PSNR of the proposed technique is 25.27%, 40% and 50.74% better than ScPSO-MT, Double Threshold and IWO-SUSAN approaches. The FSIM of the proposed technique is 28.57%, 34.12% and 34.12% better than ScPSO-MT, Double Threshold and IWO-SUSAN methods. [ABSTRACT FROM AUTHOR]

With the popularization of the Internet of Things technology and the improvement of 5G communication technology, the influence of mobile devices on the network structure is increasing. The devices in the network are usually regarded as social users that transmit information. Because the movement of users is dynamic and random, it is more difficult for complex networks to grasp the changing rules of their topological structure. The data transmission model established by considering only the historical behavior of users can no longer meet the demand for fast transmission of large-capacity data. Based on this, this paper proposes a dynamic personalized data transmission model (GRDPS) that considers the recurrent neural network and attention mechanism. First, it uses a recurrent neural network to build users' personalized preferences and model the user's historical behavior. Then, GRDPS introduces an attention mechanism to dynamically weight historical user behaviors based on the user's current message transmission. It is different from the previous methods of modeling user historical behaviors. Based on the requirements of user dynamics, GRDPS effectively considers the temporal characteristics of user historical behaviors and automatically learns the evolution law of user behaviors. Based on the demand of user randomness, GRDPS fully considers the characteristic correlation between the user's historical behavior and current transmission demand. Finally, GRDPS combines these two points to obtain a personalized ranking of users. The simulation results show that the delivery rate of GRDPS is up to 0.95. Moreover, its data transmission delay and network overhead are better than other methods in the experiment. [ABSTRACT FROM AUTHOR]

Face recognition is difficult due to the higher dimension of face image features and fewer training samples. Firstly, in order to improve the performance of feature extraction, we inventively construct a double hierarchical network structure convolution neural network (CNN) model. The front-end network adopts a relatively simple network model to achieve rough feature extraction from input images and obtain multiple suspect face candidate windows. The back-end network uses a relatively complex network model to filter the best detection window and return the face size and position by nonmaximum suppression. Then, in order to fully extract the face features in the optimal window, a face recognition algorithm based on intermediate layers connected by the deep CNN is proposed in this paper. Based on AlexNet, the front, intermediate and end convolution layers are combined by deep connection. Then, the feature vector describing the face image is obtained by the operation of the pooling layer and the full connection layer. Finally, the auxiliary classifier training method is used to train the model to ensure the effectiveness of the features of the intermediate layer. Experimental results based on open face database show that the recognition accuracy of the proposed algorithm is higher than that of other face recognition algorithms compared in this paper. [ABSTRACT FROM AUTHOR]

The image generation and completion model complement the missing area of the image to be repaired according to the image itself or the information of the image library so that the repaired image looks very natural and difficult to distinguish from the undamaged image. The difficulty of image generation and completion lies in the reasonableness of image semantics and the clear and true texture of the generated image. In this paper, a Wasserstein generative adversarial network with dilated convolution and deformable convolution (DDC-WGAN) is proposed for image completion. A deformable offset is added based on dilated convolution, which enlarges the receptive field and provides a more stable representation of geometric deformation. Experiments show that the DDC-WGAN method proposed in this paper has better performance in image generation and complementation than the traditional generative adversarial complementation network. [ABSTRACT FROM AUTHOR]

This paper is concerned with the global asymptotic stability (GAS) problem of fixed-point Lipschitz nonlinear digital filters employing 2's complement overflow arithmetic. Nonlinear digital filtering finds immense applications in various fields such as adaptive systems and controllers, digital controllers and observers for nonlinear systems, realization of neural networks using digital hardware, controllers for feedback linearization, etc. Lipschitz nonlinear digital filter is considered in this paper as it is frequently employed in nonlinear digital filtering, state filtering, neural networks, feedback control, digital controllers, decision-taking systems and so on. Based on Lyapunov theory, the property of 2's complement overflow arithmetic and Lipschitz condition associated with system nonlinearities, a new criterion for the suppression of overflow oscillations in 2's complement state variable realizations of digital filters is established. Several examples along with simulation results are provided to highlight the utility of the criterion. [ABSTRACT FROM AUTHOR]

Advanced Persistent Threat (APT) is a multi-stage and multi-step attack process. The reconstruction of the APT attack scene can start with discrete stage attack detection. However, due to the strong characteristic of concealment of APT attacks, some discrete events in the attack scenarios may not be detected. Therefore, to reconstruct the APT attack scene, we need to mine the hidden attack events according to the APT attack target and the detected discrete attack events, describe the action sequence according to the time sequence or the conditions reached by the attack, and finally reconstruct the attack path. In this paper, we depend on the EP-IKC attack cooperation model, we take the total target of APT attack as the pyramid vertex, and the alerted network entities and potential attacked entities related to the vertex as the facet nodes, this paper introduces the hidden Markov model (HMM), and uses the methods of data association and advanced probability theory to mine the hidden APT attack stages, Finally, the detection of APT attack process and the reconstruction of attack scene are realized. [ABSTRACT FROM AUTHOR]

Cloud computing can improve the calculation and data storage ability for the control center in the power system. A new framework of the cloud-based control center is proposed in this paper. This cloud computing system can collect the load data from smart meters of the grid and classify demand-side management (DSM) loads that meet the specific requirements. The selected loads belong to the off-peak period (from 21:00 to 07:00 next day) and can contribute to shifting the night peak load. A feature extraction combined with Variational Mode Decomposition (FE-VMD) of the loads which can be trained in recurrent neural network (RNN) is proposed in this paper. Using the feature value to replace the actual load data, input data can be significantly reduced which is suitable for a vast amount of load in the power system. A case study of real load data from 200,000 customers has been classified with this method, and the accuracy is compared with the other methods. From simulation with MATLAB, it can be seen that the FE-VMD combined with the RNN method provides the best result of 89.8% recognition accuracy among these methods. [ABSTRACT FROM AUTHOR]

Crowdsourcing design is a knowledge-intensive innovative design mode, in which there are various intellectual property risks. This paper proposes a novel multi-criteria decision-making approach for intellectual property risk assessment in crowdsourcing design. First, domain experts are invited to identify and evaluate various risk factors in the given crowdsourcing design environment, and the interval-valued Pythagorean fuzzy sets theory (IVPF) is used to eliminate the vagueness and ambiguity of risk assessment opinions from experts. A combination weight method based on the Decision Making Trial and Evaluation Laboratory (DEMATEL) and entropy weight (EW) is designed to measure the importance degree of risk evaluation parameters, which includes risk probability (P), risk frequency (F) and risk severity (S). Then, a TODIM method is used to calculate the risk value of various factors to identify the largest intellectual property risk. Finally, the feasibility and effectiveness of the proposed approach are verified by a case. The experimental results show that the proposed approach can effectively evaluate the intellectual property risks in crowdsourcing design. [ABSTRACT FROM AUTHOR]

The principal aim of Sustainable Development Goals (SDGs) or Global Goals is to attain a much better and more sustainable future for all. In recent times, microgrids have attracted much attention, given the transformation of software systems and raising the bar for customers in terms of sustainability, reliability and cost predictability. Consequently, there is a need for microgrid development in remote areas. Our paper proposes a hybrid renewable energy microgrid (HREM) that focuses on the affordable, efficient, reliable and sustainable growth of energy systems. This paper presents an optimized off-grid HREM for a remote locality in the south Indian state of Tamil Nadu. The proposed approach employs a configuration of photovoltaic (PV) arrays, AC loads, a diesel generator set, a wind turbine and a battery energy storage system (BESS) connected to AC/DC buses and designed to satisfy the power requirements of a remote, rural community residential area. The primary objective was to carry out a cost optimization of the designed system. A sensitivity analysis was also conducted to obtain a feasible solution from the optimized results and evaluate the robustness of the design. The primary objective was to carry out a cost optimization of the designed system. A sensitivity analysis was also conducted to obtain a feasible solution from the optimized results and evaluated the robustness of the design. A detailed analysis was made by comparing the PV + DG set + battery and PV + wind + DG set + battery and it was found that the former offered better results for the particular location. Overall, it was observed that our proposed system offers renewable energy for residential loads that is relatively inexpensive, reliable and sustainable. [ABSTRACT FROM AUTHOR]

This paper presents the design procedure of a high-performance sensorless control strategy for the widely used brushless DC (BLDC) motors. Generally, conventional sensorless techniques are based on detecting the zero-crossing instants (ZCP) of the back electromotives forces (back-EMFs) of the three phases. These methods, although widely adopted and marketed on an industrial level, involve many limitations such as filtering delays, difficulty to operate at low speeds and immunity against Electromagnetic Interferences (EMI). In this paper, the main objective is to develop a sensorless control technique integrally independent from the zero-crossing points of the back-EMFs. In the proposed method, a zero-delay adaptive filter was used to extract the fundamental and the quadrature components of the line-to-line voltage of the motor. Then, the Synchronous Reference Frame Phase Locked Loop (SRF-PLL) is used to estimate the electrical angle of phase-to-phase back-EMF along with the rotor speed. The conventional SRF-PLL was implemented using a second-order loop filter (type-3 PLL) in order to improve synchronization performances and for better rejection of voltage spikes induced in motor phases during commutations. The benefits of the control technique are brought to light through simulation results. An experimental prototype was designed to confirm the validity of the proposed method. [ABSTRACT FROM AUTHOR]

In recent years, solar photovoltaic-based power generation technology has become the key planning direction of many countries around the world. In the process of making solar cells, the quality inspection requirements are very particular, such as physical damages, surface scratches, broken grids and microcracks. In traditional factory production, the detection of the above defects requires professional inspectors to carry out visual inspection, which often leads to low detection efficiency, subjective assumption and fatigue, as well as some detection errors. In recent years, the rapid development of computer vision makes it possible to be used to detect the defects in solar cells. To overcome existing barriers, this paper proposes a method for detecting surface defects in solar cells based on deep neural network. Specifically, a specified image segmentation model named U-Net is developed for this purpose. By automatically segmenting little objects using the proposed recognition approach, surface defects detection can be realized. At last, we use a set of experiments on images from real scenes to verify the proposed method. [ABSTRACT FROM AUTHOR]

GPS receivers have a wide range of applications, but are not always secure. A spoofing attack is one source of conscious errors in which the counterfeit signal overcomes the authentic GPS signal and takes control of the receiver's operation. Recently, GPS spoofing attack detection based on computational algorithms, such as machine learning, classification, wavelet transform and clustering, has been developing. This paper proposes multiple clustering algorithms for accurately clustering the authentic and spoofing signals, called subtractive, FCM and DBSCAN clustering. The spoofing attack is recognized using two distinct features: moving phase detector variance and norms of correlators. Spoofing and authentic signals have different patterns in the proposed features. According to the Dunn and Silhouette indexes, the validation of the results is investigated. The Dunn values for the proposed approaches are 0.8592, 0.5285 and 0.6039 for DBSCAN, FCM and subtractive clustering, respectively. Also, the DBSCAN algorithm is implemented at the RTL level because of its highest value for the Dunn index and algorithm verifiability. Using the Vivado tools, this algorithm is implemented and designed on a Xilinx Virtex 7 xc7vx690tffg1930-3 hardware device for two-dimensional data with 32-bit accuracy and 130 data points. [ABSTRACT FROM AUTHOR]

This paper presents the implementation of a sense amplifier for a low-power cardiac pacemaker using the Differential Voltage Current Conveyor (DVCC). Two significant aspects of the pacemaker are sensing and pacing. The pulse generator, which is the heart of the pacemaker, consists of a sense amplifier, a logic unit and a timing control unit. The sense amplifier comprises an instrumentation amplifier, a bandpass filter and a comparator that are used to detect the QRS complex wave from the cardiac signal. Based on the output of a sense amplifier, the logic unit and the timing control unit decide whether to pace the heart or not, which achieves the requirement of the demand pacing. In this paper, a novel design of the sense amplifier using a DVCC is proposed, and the simulations are performed using 130-nm TSMC technology. Furthermore, the modes of the pacemaker VVI, DDD and rate-responsive algorithms have been implemented using the structural approach in VHDL by taking into consideration the timing cycles of a pacemaker. The design analysis shows that the proposed model of pacemaker is highly efficient and consumes significantly less energy. [ABSTRACT FROM AUTHOR]

The aim of this paper is to address the analysis and design problem of H ∞ filtering for discrete time-varying delay systems with missing measurements. Our attention is focused on the filter design, using Scaled Small Gain (SSG) approach, guaranteeing the asymptotic stability of the augmented system. A transformation model is obtained via three-term approximation method and Input–Output (IO) approach based on the SSG theorem. The use of SSG theorem for the stability of discrete time-varying delayed systems with missing measurements has not been studied elsewhere in the literature. This represents the main innovation of this paper. Less conservative results are obtained, compared with previous results, and are established in terms of LMIs. Numerical examples illustrate the applicability of the proposed methodology. [ABSTRACT FROM AUTHOR]

This paper introduces the use of an extra X current conveyor (EXCCII) for realizing logic operations, suited for mixed signal design, without necessitating separate digital blocks. The NOR and OR logic functions are thus shown to be realized using the current-mode analog building block operable at a low voltage. The use of ± 0. 7 5 V supply makes the new circuits operable at low voltage. The EXCCII bias current used in the design is 25 μ A. The input currents for defining logic 0 and logic 1 are, respectively, taken as 0 mA and 0.5 mA. The proposed circuits' delay is found to be 10 ns for low-to-high transition, while 2 ns for high-to-low transition for NOR gate under capacitive loading conditions. The use of the current-mode approach makes the new proposed circuit apt for current-mode signal processing. The new proposed logic function circuits are compatible with analog circuits, thus providing an easy merger for mixed signal systems design, paving the way to a modular design approach. As an extended application in artificial neural networks, the proposed circuit is shown to generate a sigmoid activation function with convincing results. Several results are included by varying the reference current, and the circuit's robustness aspect is also tested in the presence of fluctuation in the bias current of EXCCII. The novel approach of utilizing an analog building block for realizing digital functions is, therefore, verified with promising future applications. [ABSTRACT FROM AUTHOR]

In this paper, a novel system for measuring bio-potentials, including electroencephalography (EEG), electrocardiography (ECG) and electromyography (EMG) signals, was implemented. This system is based on the high-precision (24-bit) analog front-end ADS1299 with eight input channels. The aim of this work is to provide a low-cost platform for researchers in neuroscience, brain–computer interfaces, ECG pattern recognition and myoelectric control for Robotic Hand-Assisted Training, etc. Compared to the existing systems, this design uses a module called ESP-WROOM-32 based on a 32-bit dual-core Xtensa LX6 microprocessor in which all control and communication functions have been integrated into a single package, giving the possibility to interface the system with the Raspberry Pi via the USB interface or via the wireless interface (Wi-Fi and Bluetooth). The paper presents a detailed study of the system in terms of hardware and software implementation. In addition, an experimental process has been conducted with the aim of evaluating the proposed prototype. With a common mode rejection ratio higher than 110 dB and an input referred noise less than 2 μ V (peak-to-peak) as well as the good quality of the measured biopotentials during all the proposed scenarios, the model can be qualified to be functioning properly following the recommendations of the ADS1299 manufacturer. Finally, a conclusion is made to summarize the results achieved while highlighting the future study and the suggestions for improving the presented design. [ABSTRACT FROM AUTHOR]

Mobile edge computing (MEC) is one of the efficient technologies to provide satisfying quality of experience (QoE) for emerging computation-intensive applications in internet of things (IoT). However, some new challenges will be encountered when MEC is applied in a large-scale IoT with massive access devices or heavy traffic loads such as load balancing and traffic offloading. Aiming at the solution of these problems, this paper proposes a learning-based traffic control architecture for IoT with MEC. Moreover, a deep-learning-based load balancing framework is developed to control user association in IoT. The user association is determined at each IoT access points (IAP) by the deep neural network (DNN), which is the duplication of the well-trained DNN with the global network information. In addition, we propose a reinforcement-learning-based partial traffic offloading scheme to reduce the traffic origination. The IoT devices are able to independently decide its offloading radio according to the channel quality information, service requirement, and workload of the IAP. Simulation results indicate that the proposed deep-learning-based load balancing scheme is able to achieve uniform traffic distribution, and meanwhile our partial offloading policy can significantly reduce the network traffic. [ABSTRACT FROM AUTHOR]

The negative solar resistance, is highly dependent on the solar operating point. The system will become sluggish and unstable during low irradiance. It is reported that the impact of negative solar resistance is nullified by parallel impedance emulated condition (PIEC). During low irradiance, the system stability has been moved to an unstable position which is not fully resolved by PIEC. In this paper, a virtual inertia-damping voltage and current solar droop emulated controller is proposed to improve the voltage oscillation profile, inertia problem and damping of the DC microgrid system. In DC microgrids, the inherent inertia from DC capacitors is low, thus affecting the bus voltage oscillation during a reduction of irradiance. The oscillation of the system has been increased due to the equivalent impedance by negative solar resistance at lower irradiance. Oscillation's issue and voltage instability of the PV system have been enhanced by the inner current controller. In addition, the steady-state error of the PV voltage is minimized by the virtual inertia-damping voltage restoration droop controller with PIEC. Performance analysis and dynamic characteristics are discussed through a pole-zero plot for the various conditions of negative solar resistance using virtual inertia-damping voltage and current Solar Droop Emulated Controller. This paper focuses on stability analysis to strengthen the DC microgrid. The performance of the solar controller with bidirectional converter (BDC) controller has been discussed to verify the power management between the system to manage the power for the unbalanced load. [ABSTRACT FROM AUTHOR]

In this paper, a differential-input plus-type differential voltage current conveyor-based first-order universal filter is designed. This voltage-mode filter uses a grounded capacitor. In addition, it can provide all the noninverting and inverting first-order universal filter responses. The circuit provides high common-mode rejection ratio of about 76.5 dB. Nevertheless, it needs a single matching problem and comprises two floating resistors. Quadrature oscillator (QO) design is obtained using this filter as an application example. The designed filter and QO circuits are simulated through the SPICE program, and some experimental studies are carried out using AD844 ICs to verify the theory. [ABSTRACT FROM AUTHOR]

In recent years, virtual reality (VR) has gradually entered the daily education and teaching activities from pure scientific research. In the area of assistance teaching, some typical computer softwares still play some important roles. This makes remote teaching activities can just learn voice, yet cannot possess the feeling of realistic existence. Especially in scenario of COVID-19, remote teaching activities with proper perceptibility are in urgent demand. To deal with the current challenge, this paper proposes a wireless VR-based multimedia-assisted teaching system framework under mobile edge computing networks. In this framework, cooperative edge caching and adaptive streaming based on viewport prediction are adopted to jointly improve the quality of experience (QoE) of VR users. First, we investigated the resource management problem of caching and adaptive streaming in this framework. Considering the complexity of the formulated problem, a distributed learning scheme is proposed to solve the problem. The experimental data are verified and the experimental results prove that the studied methods improve the performance of user QoE. [ABSTRACT FROM AUTHOR]

The popularization of smart mobile devices has led to the research on wireless power transfer (WPT) technology. Previous research shows that increasing the excitation coil or the energy transfer coil can increase the energy transmission distance of the WPT system. But the mutual inductance calculation of coils is not accurate enough when the distance is close. This paper presents an improved formula to calculate the mutual inductance between two round inductance coils. Contrasts with the previous calculation methods, computer simulation and experimental results are done to validate the accuracy of the formula, especially when the two inductance coils are close together. On the basis of this formula, the maximum output power (MOP) and the maximum output power distance (MOPD) of the two-, three- and four-coil series–series resonance WPT systems based on circuit models are studied and compared. The simulation and experimental results show a high consistency with the theoretical calculation results. We can find that the farther the MOPD of the system is, the smaller the MOP is. Moreover, when the operating frequency increases, the MOPD of the two-coil conformation increases, while the four-coil conformation is just opposite. [ABSTRACT FROM AUTHOR]

Energy harvesting from the microbial fuel cells have a significant attention in the recent days, due to their cost efficiency, simple designing structure and self-powered system. Also, the emergence of internet of things plays a vital role in many real time application scenarios like agricultural purposes and activities. But, the incorporation of these techniques is one of challenging and interesting tasks in the research field. In the conventional works, the internet of things has been utilized as a cloud storage domain for activating the sensors used for environmental monitoring and controlling purposes. The main intention of this paper is to design a robust and cost-effective sludge water based microbial fuel cells, and utilize it for an internet of things incorporated ecological monitoring and farming applications by activating the smart sensors. It discusses about the various electrode combination with several mixture of substrate to study about the optimum performance of microbial fuel cells. To ease the comparative study, Thing Speak platform is used along with the necessary sensors for continuous monitoring. In addition to that, the efficiency of single and dual chamber microbial fuel cell is analyzed based on the set of parameters such as cost, size, and construction. In this work, the microbial fuel cell-based energy harvesting scheme is also developed with switched capacitance-based metal oxide semiconductor field effect transistor and relay-based charge pump circuit which can be incorporated to the internet of things based agriculture applications. Here, the cost analysis of microbial fuel cell with and without DC–DC converter have been compared for selecting the most suitable one for the application system. Moreover, the digital temperature and humidity sensor can be utilized with the proposed microbial fuel cell system for gathering the inputs of the ecological system, which acts as an interface of the microbial fuel cell and cloud systems. During experimentation, the results of both the energy harvesting schemes are evaluated and compared by using various performance indicators. [ABSTRACT FROM AUTHOR]

With the rapid development of advanced driver assistance systems (ADASs) and infotainment systems, automotive networks have higher requirements for bandwidth and real-time than before. Time-triggered Ethernet (TTE) supports real-time time-triggered communication and event-triggered communication coexistence. TTE is used as a backbone network in intelligent connected vehicles to meet the growing real-time and reliability requirements. To design a real-time TTE switch, the time delay of data streams needs to be analyzed. Most of the recent works analyze end-to-end latency, which is usually based on a particular integration policy and ignores the impact of BE flows. In this paper, we analyze the delay bounds of RC flows in the TTE switch using delay reachability techniques. We consider the impacts of TT flows, RC flows and BE flows on RC flows based on three integration policies: shuffling, preemption and timely blocking, respectively. We obtain a secure upper limit of delay through analysis. Experiments illustrate the degree of influence of each factor on the delay bound of RC flow, which provides a reference for the optimal design of the TTE switch in intelligent connected vehicles. [ABSTRACT FROM AUTHOR]

As the latest generation of cellular mobile communication technology, the fifth-generation (5G) mobile communication technology is proposed to build a network-based society. As a result, industry and academia are placing great emphasis on security research for 5G technology. 5G technology has higher security requirements than previous systems and the secondary authentication is an important technology to ensure the security of 5G. In this paper, we propose to incorporate the SM9 encryption algorithm with 5G secondary authentication to provide higher security assurance. Due to the high computational overhead of the SM9 algorithm, we further propose an acceleration strategy for the SM9 algorithm based on the non-adjacent form (NAF) representation method to improve the performance of the entire system. [ABSTRACT FROM AUTHOR]

A radiation-hardened delay-locked loop (DLL) applied for multiple frequency ranges is proposed in this paper. Novel and effective radiation-hardened techniques are applied in charge pump (CP) and voltage-controlled delay line (VCDL) to avoid inverted lock error and missing pulses error caused by single-event transient (SET). In order to ensure a wide operation frequency range, a frequency detection module is designed to detect the frequency of input signals and adjust the current source array of VCDL so as to change its delay time. Besides, an accelerate module is designed to reduce the lock time of DLL by providing an extra discharge current. According to the simulation results, the proposed DLL shows good performance when ion strikes occur and can work properly in a wide operation frequency range from 3.7 GHz to 7.4 GHz. Also, the proposed accelerate module helps to reduce lock time by 22.2%. [ABSTRACT FROM AUTHOR]

The class-imbalance learning is one of the most significant research topics in the data mining and machine learning. Imbalance problem means that one of the classes has much more samples than that of other classes. To deal with the issues of low classification accuracy and high time complexity, this paper proposes an novel imbalance data classification algorithm based on clustering and SVM. The algorithm suggests under-sampling in majority samples based on the distribution characteristics of minority samples. First, specific clusters are detected by cluster analysis on the minority. Second, a cluster boundary strategy is proposed to eliminate the bad influence of noise samples. To structure a balanced dataset for imbalance data, this paper proposes three principles of under-sampling on majority samples according to the characteristic of samples in the cluster. Finally, the optimal classification model from the linear combination of hybrid-kernel SVM is obtained. The experiments based on datasets in UCI and KEEL database show that our algorithm effectively decreases the interference of noise samples. Compared with the SMOTE and Fast-CBUS, the proposed algorithm not only reduces the feature dimension, but also improves the precision of the minor classes under the different labeled sample rates generally. [ABSTRACT FROM AUTHOR]

Accurate ramp signal, with low power dissipation, is highly demanded, for applications like counter ADC. This paper presents a novel low power ramp generator circuit with a negative feedback loop for compensation of the variations in process, voltage, and temperature (PVT). While using an opamp for PVT compensation has been essential in the previous ramp generator structures, the proposed ramp generator is opamp-less. Derived equations of the proposed ramp generator circuit show that PVT compensation structure works effectively. In addition, the circuit design and simulations were done in TSMC 0.18 μ m CMOS technology. Corner analysis shows that integral non-linearity (INL) of the ramp signal is about 3.7 mV, for a wide temperature range, while the power dissipation of the circuit is about 1.16 μ W. [ABSTRACT FROM AUTHOR]

Existing block-level defect detection method in patterned fabric causes a large number of false detections due to the lack of edge information. To solve this problem, in this paper, we propose a bilayer Markov random field (BMRF) method for inspecting defects in patterned fabric. First, the proposed method reduces samples of the original fabric image to obtain the constraint layer, which can locate the defective block roughly. Second, we interpolate samples into the image to supplement the local information to improve and optimize the imperfect boundary, to obtain a more detailed data layer. Moreover, this paper proposes a new potential function, which considers the differential characteristics of the image blocks in the same layer and the transition probability between different layers. Finally, this paper utilizes a parameter estimation method based on the expectation maximization to solve the parameters of the BMRF method. The proposed BMRF method is evaluated on databases of star-, box- and dot-patterned fabrics. By comparing the resultant and ground-truth images, the recall rate of the proposed method in the three patterned fabrics is 95.32%, 89.29% and 93.28%, respectively, which is comparable to the existing methods. [ABSTRACT FROM AUTHOR]

This paper introduces an FPGA implementation of a pseudo-random number generator (PRNG) using Chen's chaotic system. This paper mainly focuses on the development of an efficient VLSI architecture of PRNG in terms of bit rate, area resources, latency, maximum length sequence, and randomness. First, we analyze the dynamic behavior of the chaotic trajectories of Chen's system and set the parameter's value to maintain low hardware design complexity. A circuit realization of the proposed PRNG is presented using hardwired shifting, additions, subtractions, and multiplexing schemes. The benefit of this architecture, all the binary multiplications (except X i ⋅ Y i and X i ⋅ Z i) operations are performed using hardwired shifting. Moreover, the generated sequences pass all the 15 statistical tests of NIST, while it generates pseudo-random numbers at a uniform clock rate with minimum hardware complexity. The proposed architecture of PRNG is realized using Verilog HDL, prototyped on the Virtex-5 FPGA (XC5VLX50T) device, and its analysis has been done using the Matlab tool. Performance analysis confirms that the proposed Chen chaotic attractor-based PRNG scheme is simple, secure, and hardware efficient, with high potential to be adopted in cryptography applications. [ABSTRACT FROM AUTHOR]

Adversarial attack on neural networks has become an important problem restricting its security applications, and among adversarial attacks oriented towards the sample set, the universal perturbation design causing most sample output errors is critical to the study. This paper takes the neural network for image classification as the research object, summarizes the existing universal perturbation generation algorithm, proposes a universal perturbation generation algorithm combining batch stochastic gradient rise and spherical projection search, achieves loss function reduction through the iterative training of stochastic gradient rise in batch samples, and limits the universal perturbation search to a high-dimensional sphere with radius ε to reduce the search space of universal perturbation. Moreover, the regularized technology is introduced to improve the generation quality of universal perturbations. The experimental results show that compared with the baseline algorithm, the attack success rate increases by more than 10%, the solution efficiency of universal perturbation is improved by one order of magnitude, and the quality controllability of universal perturbation is better. [ABSTRACT FROM AUTHOR]

The wind power interval prediction of offshore wind farms and power plan arrangement of conventional thermal power units are of vital importance in the consumption of offshore wind power, the reduction of greenhouse gas impact on the environment, and the electric power system safe and economic operating. With the purpose of selecting the appropriate Copula function on the basis of the results of wind speed and wind power normal test, establish the mathematical model of wind-fire joint optimal scheduling, and optimize coal-fired power units power generation after comparing the convergence performance of particle swarm optimization method and crow search algorithm. Results indicate that the selected Copula function meets the expected criteria, and the optimized thermal unit climbs more smoothly and through the optimization of CSA the complete economic consumption of running is lessened. An idea is presented by this paper, which considers the uncertainties of offshore wind power generation, and the basis for the operational performance of CSA over PSO, and which provides a joint wind-thermal economic optimal dispatch strategy. [ABSTRACT FROM AUTHOR]

Machine learning (ML) can be used for deep mining and analysis of multidimensional medical data. At present, it has been widely used in medical diagnosis and prognosis prediction. This paper aims to make the existing research no longer focus on identifying key risk factors of stroke, and predict stroke risk more accurately. We collected the data of 3,962 cerebral apoplexy patients from 2019 to 2020, according to gender (male: 2,613; female: 1,349) and age (16–40 years old; 41–54 years old; 55–69 years old; 70 years old and above) layered. After data preprocessing, a stroke risk prediction model was built using principal component analysis (PCA) and extreme learning institutions (ELM). The prediction accuracy of PCA-ELM was as high as 97%. In this model, total cholesterol and high density lipoprotein are taken as 10 important factors that affect the incidence of stroke. This method can timely and efficiently mine the factors influencing the incidence of cerebral apoplexy from the data, and can predict the incidence of cerebral apoplexy. It has high value in practical application. This paper has great reference value in the research of brain death. [ABSTRACT FROM AUTHOR]