A novel interpretable multilevel wavelet decomposition deep network for actual heartbeat classification

Arrhythmias may lead to sudden cardiac death if not detected and treated in time.A supraventricular premature beat(SPB)and premature ventricular contraction(PVC)are important categories of arrhythmia disease.Recently,deep learning methods have been applied to the PVC/SPB heartbeats detection.However,most researchers have focused on time-domain information of the electrocardiogram and there has been a lack of exploration of the interpretability of the model.In this study,we design an interpretable and accurate PVC/SPB recognition algorithm,called the interpretable multilevel wavelet decomposition deep network(IMWDDN).Wavelet decomposition is introduced into the deep network and the squeeze and excitation(SE)-Residual block is designed for extracting time-domain and frequency-domain features.Additionally,inspired by the idea of residual learning,we construct a novel loss function for the constant updating of the multilevel wavelet decomposition parameters.Finally,the IMWDDN is evaluated on the Third China Physiological Signal Challenge Dataset and the MIT-BIH Arrhythmia database.The comparison results show IMWDDN has better detection performance with 98.51%accuracy and a 93.75%F1-macro on average,and its areas of concern are similar to those of an expert diagnosis to a certain extent.Generally,the IMWDDN has good application value in the clinical screening of PVC/SPB heartbeats.

Research on Key Issues of Consistency Analysis of Vehicle Steering Characteristics

Given the global lack of effective analysis methods for the impact of design parameter tolerance on performance deviation in the vehicle proof-of-concept stage,it is difficult to decompose performance tolerance to design parameter tolerance.This study proposes a set of consistency analysis methods for vehicle steering performance.The process of consistency analysis and control of automotive performance in the conceptual design phase is proposed for the first time.A vehicle dynamics model is constructed,and the multi-objective optimization software Isight is used to optimize the steering performance of the car.Sensitivity analysis is used to optimize the design performance value.The tolerance interval of the performance is obtained by comparing the original car performance value with the optimized value.With the help of layer-by-layer decomposition theory and interval mathematics,automotive performance tolerance has been decomposed into design parameter tolerance.Through simulation and real vehicle experiments,the validity of the consistency analysis and control method presented in this paper are verified.The decomposition from parameter tolerance to performance tolerance can be achieved at the conceptual design stage.

Resource Allocation for Physical Layer Security in Heterogeneous Network with Hidden Eavesdropper

The tremendous performance gain of heterogeneous networks(Het Nets) is at the cost of complicated resource allocation. Considering information security, the resource allocation for Het Nets becomes much more challenging and this is the focus of this paper. In this paper, the eavesdropper is hidden from the macro base stations. To relax the unpractical assumption on the channel state information on eavesdropper, a localization based algorithm is first given. Then a joint resource allocation algorithm is proposed in our work, which simultaneously considers physical layer security, cross-tier interference and joint optimization of power and subcarriers under fairness requirements. It is revealed in our work that the considered optimization problem can be efficiently solved relying on convex optimization theory and the Lagrangian dual decomposition method is exploited to solve the considered problem effectively. Moreover, in each iteration the closed-form optimal resource allocation solutions can be obtained based on the Karush-Kuhn-Tucker(KKT) conditions. Finally, the simulation results are given to show the performance advantages of the proposed algorithm.

THE ANALYSIS OF THE SEPARATED LAYERS ALGORITHM BY DOMAIN DECOMPOSITION METHOD

From the principle of of the Domain Decomposition Method (DDM), we analyse the 2nd-order linear elliptic partial differential problems and link the Separated-Layers Algorithm (SLA) with DDM. The mathematical properties of SLA and numerical example are presented to obtain satisfactory computation results. For general linear differential ones, also are the structure of SLA and its characteristics discussed.

Electron transport layer driven to improve the open-circuit voltage of CH_3NH_3PbI_3 planar perovskite solar cells

Suitable electron transport layers are essential for high performance planar perovskite heterojunction solar cells. Here, we use ZnO electron transport layer sputtered under oxygen-rich atmosphere at room temperature to decrease the hydroxide and then suppress decomposition of perovskite films. The perovskite films with improved crystallinity and morphology are achieved. Besides, on the ZnO substrate fabricated at oxygen-rich atmosphere, open-circuit voltage of the CH_3NH_3PbI_3-based perovskite solar cells increased by 0.13 V.A high open-circuit voltage of 1.16 V provides a good prospect for the perovskite-based tandem solar cells. The ZnO sputtered at room temperature can be easily fabricated industrially on a large scale, therefore, compatible to flexible and tandem devices. Those properties make the sputtered ZnO films promising as electron transport materials for perovskite solar cells.