Source localization is a key technology in array signal processing. However, it has a serious problem that the direction of arrival (DOA) estimation accuracy is reduced at low signal-to-noise ratio (SNR). Thus, this Letter proposes an improved near-field multiple signal classification (INF-MUSIC) algorithm to improve DOA estimation accuracy at low SNR. Firstly, the fourth-order cumulant is used to construct a Hermitian matrix with only DOA information. Secondly, the spatial spectrum has physical property that is a breakpoint and approaches a larger value at the DOA, thus first derivative of spatial spectrum is used to improve DOA estimation accuracy at low SNR. Finally, the corresponding ranges are estimated one by one by the onedimensional MUSIC algorithm. The simulation results show that the INF-MUSIC algorithm makes the DOA estimation accuracy improved by 2° at low SNR.

Electrocardiogram (ECG) signals are often used to diagnose cardiac status. However, most of the existing ECG diagnostic methods only use the time-domain information, resulting in some obviously lesion information in frequency-domain of ECG signals are not being fully utilized. Therefore, we propose a method to fuse the time and frequency domain information in ECG signals by convolutional neural network (CNN). First, we adapt multi-scale wavelet decomposition to filter the ECG signal; Then, R-wave localization is used to segment each individual heartbeat cycle; And then, the frequency domain information of this heartbeat cycle is extracted via fast Fourier transform. Finally, the temporal information is spliced with the frequency domain information and input to the neural network for classification. The experimental results show that the proposed method has the highest recognition accuracy (99.43%) of ECG singles compared with state-of-the-art methods.

In the inductive power transfer (IPT) system, it is recommended to drive the resonant inverter in zero-voltage switching (ZVS) or zero-current switching (ZCS) operation to reduce switching losses, especially in dynamic applications with variable couplings. This paper proposes an improved autonomous current-fed push-pull parallel-resonant inverter, which not only realizes the ZVS operation by tracking the zero phase angle (ZPA) frequency, but also improves the output power and overall efficiency in a wide range by reducing gate losses and switching losses. The ZPA frequencies characteristic of the parallel-parallel resonant circuit in both bifurcation and bifurcation-free regions is derived and verified by theory and experiments, and the comparative experimental results demonstrate that the improved inverter can significantly increase the output power from 7.68 W to 8.74 W and has an overall efficiency ranging from 63.5% to 72.5% compared with the traditional inverter at a 2 cm coil distance. Furthermore, with a 2-fold input voltage (24 V), the improved inverter can achieve an approximate 4-fold output power of 38.9 W and overall efficiency of 83.6% at a 2 cm coil distance.

Nowadays, accurate and fast vehicle detection technology is of great significance for the construction of intelligent transportation systems in the context of the era of big data. This paper proposes an improved lightweight YOLOX real-time vehicle detection algorithm. Compared with the original network, the detection speed and accuracy of the new algorithm have been improved with fewer parameters. First, referring to the GhostNet, we make a lightweight design of the backbone extraction network, which significantly reduces the network parameters, training cost, and inference time. Furthermore, by introducing the α-CIoU loss function, the regression accuracy of the bounding box(bbox) is improved while the convergence speed of the model is also accelerated. The experimental results show that the mAP of the improved algorithm on the BIT-Vehicle dataset can reach up to 99.21% with 41.2% fewer network parameters and 12.7% higher FPS than the original network and demonstrate the effectiveness of our proposed method.

Based on the rapid development of network technology, from the perspective of constructing an OMO (Online-Merge-Offline) learning space and combining virtual and real experimental teaching modes, this paper mainly studies the practical application method of the OMO education model. The integration of various teaching functions in the OMO learning space provides learners with more convenient services and allows teaching staff to implement teaching more efficiently. Relying on the virtual simulation experiment platform combining virtual and real, modern information technology and experimental teaching are deeply integrated, and a specific experimental teaching scheme is given, which better solves the problems of the limitation of the experimental platform, the single experimental teaching mode and the lack of experimental teaching resources. At the same time, the breadth and depth of experimental teaching content are expanded, the time and space of experimental teaching are extended, and the quality and level of experimental teaching are improved.