Pulmonary edema assessment is a key factor in monitoring and guiding the treatment of critically ill patients. To date, the methods available at the bedside to estimate the physiological correlation of pulmonary edema and extravascular pulmonary fluid are often unreliable or require invasive measurements. The aim of this article is to develop an imaging method of reliably assessing pulmonary edema by utilizing functional electrical impedance tomography. In this article, the Split-Bregman algorithm is used to solve the Total Variation (TV) minimization problem in EIT image reconstruction. A thorax model is constructed according to CT images of rats. Through simulation and experiment, the proposed method improves the quality of reconstructed image significantly compared with existing methods. A pulmonary edema experiment in rats is also carried out. The development of pulmonary edema is analyzed numerically through EIT images.....

The performance of the Vertical Cavity Surface Emitting Laser (VCSEL) for hybrid optical links SMF/FSO based on different data rates and MIMO configuration techniques was obtained using OptiSystem^TM which is close to the results of the experimental system. The developed system was tested with various transmission distances: 20, 30, 40, and 50 km, and in the existence of many configuration kinds and modulations. In addition to that the hybrid system was estimated with different weather cases: clear, rain, and snow. The results state that the performance of the OOK-NRZ system reveals better performance than OOK-RZ system under the same conditions. Also, the performance of the free space link is better than the fiber link formost of the link ranges considered and configurations. For OOK-NRZ of the fiber link, it was found that the MIMO 8×8 technique has better system performance than other configurations, and the Q-factor = 11.39 and BER = 5.4×10^-30 for a length of 50 Km while for the FSO link, it was found that MIMO 8×8 indicates a high performance for Q-factor = 12.7 and BER = 1.8×10^-37. The maximum FSO link distances under different weather conditions and coupling ratios were found. For BER≤10⁻⁹, in NRZ format for SMF 50 km utilizing MISO8×1 technology in clear weather for 10 Gbps, 15 Gbps, and 20 Gbps for FSO links, the maximum accessible lengths are 0.6 Km, 0.51 Km, and 0.43 Km, respectively. The process is expanded to include snow conditions for data rates of 10 Gbps, 15 Gbps, and 20 Gbps for FSO links with lengths of 0.4 Km, 0.3 Km, and 0.26 Km, respectively.....

Methods of manual analysis for infrared image and temperature detection of power transmission and transformation equipment typically have problems, such as low efficiency, strong subjectivity, easy to make mistakes and poor real-time feedback. In this paper, a high temperature anomaly detection method based on SegFormer in infrared image of power transmission and transformation equipment is proposed. Many infrared images of power transmission and transformation equipment are collected and preprocessed, and the temperature information of each infrared image is read out using the DJI sdk tool to construct the temperature data matrix. In the segmentation stage, the SegFormer network is used to segment the key parts of the power transmission and transformation equipment to obtain the mask for detection. The maximum values of the temperature data in the mask area are calculated, and the high temperature anomaly detection atthe key parts of the power transmission and transformation equipment is realized. The test results on the test set show that the overall performance of the method is the highest as compared to other methods such as FCN, UNet, SegNet, DeepLabV3+, and an automatic temperature recognition can be realized, which has important practical value for the detection of high temperature anomaly at the key parts of power transmission and transformation equipment.....

Linear Frequency Modulation (LFM) signals are widely used in radar and sonar technology. Many applications are interested in determining the source of an LFM signal. In recent years, the rapid development of machine learning has facilitated research in various fields, including signal recognition. The neural networks can extract the implicit features of the signals, which can help the system to sort and recognize the signal sources quickly and accurately. High performance of neural networks requires large amounts of high-quality labeled data. However, it is difficult and expensive to obtain a large amount of high-quality labeled data. Simultaneously, some features will be lost during data preprocessing, and feature extraction and classification tasks will be inefficient. The self supervised network is proposed in this paper for pre-training the signal waveform and fine-tuning the classification with a small amount of labeled data. The proposed method can extract more signal waveform features, save labeling costs, and has higher precision. This method can provide up to 99.7% recognition accuracy at 20 dB.....