Abstract：Mobile ad hoc networks based on directional antennas are widely used in the military field, especially for airborne networks with the requirements of easy implementation, high concealment and high confidentiality. Directional antennas have more advantages than omni-directional antennas, which have longer transmission ranges, reduce mutual interference among nodes, and are able to achieve the improvement of spatial multiplexing capability. But the introduction of directional antennas can create problems while increasing the communication distance, causing disruptive side effects to many protocol layers. Currently, researchers at home and abroad have proposed a variety of algorithms to solve these problems. Combined with the complexity of airborne networks, it studies various algorithms and protocols proposed at home and abroad for the data link layer and network layer of mobile ad hoc networks based on directional antennas. It summarizes and compares three key technologies in recent years, namely, directional neighbor discovery algorithm, medium access control protocol and directional routing protocol. Techniques and protocols that are more suitable for use in airborne networks are proposed. Finally, it provides some directions for further research.

Abstract：GNSS based landing guidance has become direction of the next generation precision landing system. Its integration to CNI system is also one of the most important research issues. This paper surveys principle and key frame of joint precision approach landing system (JPALS), and then discusses domestic status and technical basis in the GNSS based landing guidance. After that, the paper presents some basic plans on JPALS capability integration for integrated CNI systems.

Abstract：Aiming at the problems of poor detection accuracy of current heart sound recognition algorithms, a new heart sound recognition method based on hybrid convolutional neural network-support vector machine model (CNN-SVM) is proposed. In order to verify the effectiveness of the method，two kinds of normal and abnormal heart sound signal databases based on PASCAL challenge experiment data are sorted out. Through preprocessing, MFCC feature extraction and PCA dimension reduction, CNN-SVM model has been input for training. The performance of the proposed method is evaluated in terms of accuracy, recall, specificity and F score.In order to verify the effectiveness of this algorithm, the hybrid CNN-SVM model is compared with the single SVM model and the CNN model respectively.Five groups of the experimental results show that the proposed method can distinguish the two different heart sound with a high average recognition rate of 99%, which is 2.48% higher than the single CNN method. It is also higher than the single SVM algorithm.

Abstract：Predicting protein-protein interaction (PPI) in plants holds significant biological implications. This study has employed four encoding methods and a deep neural network to construct a model for predicting protein interactions. The results show that the developed PPI prediction model using the integrated approach of the protein language model Ankh with a deep neural network has achieved optimal AUPR and AUC values across three plant datasets, with its Sen and MCC values also outperforming those of four other models designed for protein interaction predictions. When tested on plant PPI datasets for rice and soybean, the proposed model has yielded AUPR scores of 0.802 5 and 0.730 1 respectively, and AUC scores of 0.956 2 and 0.950 7 respectively. These outstanding results indicate that the PPI model incorporating the protein language model Ankh can serve as a promising tool for predicting protein-protein interactions in plants.

Abstract：TSN (Time Sensitive Network) is a new type of network technology that enhances the real-time, reliability and security aspects of critical data on the basis of traditional Ethernet technology, which can meet the needs of industrial Internet, automatic driving, healthcare and other fields. Aiming at the emerging technology of TSN, this paper researches and analyzes the frame preemption technology in TSN and realizes the frame preemption circuit. As an important part of TSN network switching chip, it realizes the independent control of TSN frame preemption and other chip implementations, which has good application value in related time-sensitive fields.

Abstract：To enhance the data security of solid-state drive (SSD), an SM4 data encryption and decryption module designed to meet the secondary standard of the "Cryptography test criteria for security IC" in the SSD controller chip was introduced in this article. In addition, in order to verify the correctness of module design, an automated verification platform based on Universal Verification Methodology (UVM)design was introduced in this article. Taking design function points and code coverage as measurement indicators, the data encryption and decryption module was fully verified by the verification platform, and finally reached the tapeout delivery standard of System on Chip (SoC).

Abstract：Based on the actual design requirements of DC-DC chip circuit, this paper designs a bandgap reference circuit with an input voltage range of 2.5~15 V. Through the design of the pre-regulator circuit, the reference voltage output by the bandgap reference core is transformed into a stable current source, and the negative feedback structure is formed to provide a power supply voltage for the bandgap reference core itself, thus increasing the upper limit of the power supply voltage range. Through the voltage selection circuit, the bandgap reference core is directly powered by the power supply voltage when the power supply voltage is lower than 5 V, which broadens the lower limit of the power supply voltage range. At the same time, the pre-regulator circuit and the cascode structure in the bandgap reference core brings good power supply rejection performance to this circuit. This design is based on 0.25μm BCD process, and the schematic diagram, layout design and simulation of the circuit are completed. The results show that this design can output a stable voltage of 0.8 V in the temperature range of -55℃~125℃, and the temperature coefficient is 7.78 ppm/℃. At low frequency, the PSR reaches 159 dB, and the line sensitivity is 0.001 2%.

Abstract：In order to meet the requirements of performance, power consumption, volume, weight and localization in electronic systems, one RISC-V circuit based on system-in-package technology is designed. The circuit is based on a processor that adopts an independent instruction set architecture and domestic process, and integrates domestic peripheral circuit to realize a completely independent and controllable microsystem circuit with common control and communication interfaces. After testing and verification, all functions and performance of the circuit have reached the design indicators, effectively improved the functional density, and satisfied the requirements of miniaturization, light weight and low power consumption for electronic systems.

Abstract： Current field communication represents a wireless technology that propagates through underwater environments by conducting electric current. Conventional underwater electromagnetic wave communication encounters challenges such as significant signal attenuation, intricate antenna structures, and installation difficulties. To address these issues, this study integrates the theory of underwater current field propagation and proposes an underwater communication system based on an array of polar plates.The electric field model of the array polar plate is established utilizing the finite element software Comsol. Frequency domain analysis is employed to simulate the electric field and potential distribution of the 1×4 array cut surface. Additionally, a comparative analysis of signal reception effectiveness is conducted between 1×4 and 1×2 arrays. Building upon this, the impact of six sets of positive and negative wiring methods on signal reception strength is analyzed. Transient time-domain analysis is then applied to simulate changes in input and output waveforms.Experimental results validate that enhancing the number of array rows effectively improves communication without increasing transmit power. Moreover signal waveforms exhibit minimal alteration and maintain phase stability during transmission, thereby confirming the viability of underwater communication using arrayed polar plates. The findings of this research contribute a theoretical foundation for the development of a novel current field underwater communication system.

Abstract：The coexistence of old and new devices of different versions in the network, together with the differences in interface openness and interface protocols between different manufacturers' network management platforms bring difficulties to the effective management of power communication networks, especially the frequent occurrence of failures. Based on the core technology of converged power communication network management system of different devices, this paper analyzes the deployment environment of various existing systems, analyzes the network element management mode of different brand manufacturers' management systems and the possibility of interworking of related communication protocols, and then proposes a new power communication network management fault control strategy to avoid the occurrence of faults as much as possible. Subsequently, this paper makes the optimal treatment to the fault, which can effectively avoid the occurrence of the fault according to the experimental verification.

Abstract：To address issues such as color distortion, incomplete defogging, and computational inefficiency in image defogging, this study proposes an improved image defogging algorithm based on the enhanced AOD-Net. Initially, a residual connection is introduced into the existing convolutional module, preserving the features of the first layer in the second feature fusion layer to enhance feature extraction capabilities. Subsequently, an attention module is introduced after the third feature fusion layer to strengthen the representation of crucial features in hazy images and suppress irrelevant background interference. Finally, a novel composite loss function is employed for training. Experimental results demonstrate that the proposed algorithm achieves a 3.8 dB improvement in Peak Signal-to-Noise Ratio (PSNR) and a structural similarity (SSIM) of 93.6% on a public dataset. Compared to other defogging algorithms, this algorithm exhibits superior performance in both defogging accuracy and processing efficiency.

Abstract：Document-level event extraction generally divides the task into three subtasks: candidate entity recognition, event detection, and argument recognition. The conventional approach involves sequentially performing these subtasks in a cascading manner, leading to error propagation. Additionally, most existing models implicitly predict the number of events during the decoding process and predict event arguments based on a predefined event and role order, so that the former extraction will not consider the latter extraction results. To address these issues, a multi-task joint and parallel event extraction framework is proposed in this paper. Firstly, a pre-trained language model is used as the encoder for document sentences. On this basis, the framework detects the types of events present in the document. It utilizes a structured self-attention mechanism to obtain pseudo-trigger word features and predicts the number of events for each event type. Subsequently, the pseudo-trigger word features are interacted with candidate argument features, and parallel prediction is performed to obtain various event arguments for each event， significantly reducing model training time while achieving performance comparable to the baseline model. The final F1 score for event extraction is 78%, with an F1 score of 98.7% for the event type detection subtask, 90.1% for the event quantity prediction subtask, and 90.3% for the entity recognition subtask.

Abstract：Traditional particle swarm optimization (PSO) is easy to premature convergence and fall into local optimum. Therefore, chaotic dynamic multi swarm particle swarm optimization (CDMPSO) is proposed and applied to robot three-dimensional path planning. The chaotic mapping theory is introduced to improve the quality and distribution uniformity of the initial solution of the particle population. At the same time, the grouping parallel optimization strategy is introduced to divide the population into three sub populations by using the median clustering method according to the fitness value and iterate for real-time dynamic adjustment. Different methods are used to update the population according to the characteristics of different sub populations. Compared with traditional PSO algorithm and adaptive particle swarm optimization (APSO) algorithm in MATLAB software, the improved CDMPSO algorithm has larger global search range, fewer times of falling into local optimum and shorter final path, which verifies that the improved algorithm is feasible.

Abstract：Rice is a globally important staple crop, and the accurate prediction of rice yield plays a significant role in agricultural development. Due to the influence of external environmental factors and the growth mechanisms of rice, rice yield often exhibits nonlinear characteristics, making it challenging to make precise predictions. Therefore, the CE-CGRU rice yield prediction model is proposed, which extracts features using the Copula Entropy (CE) method for nonlinear environmental factors and combines them with CNN and GRU technologies. The aim is to identify crucial features for yield prediction under specific rice varieties.Based on the analysis and performance comparison using real data from Lin'an District of Zhejiang Province, the proposed model is compared to five other yield prediction models: MLR, RF, LSTM, GRU, and CNN-LSTM. The results indicate that the CE-CGRU model achieves a MAE of 0.677, a MSE of 0.87, and a MAPE of 5.029%, demonstrating its superior capability in capturing the complex nonlinear relationship between rice yield and environmental factors. Furthermore, a comparison and analysis of different feature selection methods and time steps are conducted.

Abstract：To satisfy the future demand for multi-target tracking in hemispherical coverage, a 32-element circular polarized semi-spherical conformal phased array antenna is present in this paper. Hemispherical and cylinder configuration is adopted to meet the communication requirements at low elevation angel. Activate corner domain and different element distributions are compared in detail to decide the optimum performance. Simulated results show that the gain is higher than 13.8 dBi and the gain flatness is less than 1.8 dB within the scanning range of 0°~90°, which indicates this spherical phased array antenna can be applied in the occasions of multi-target tracking in hemispherical coverage.

Abstract：With the continuous development of microelectronic components design capability and process technology, the reliability and life of the device is increasing. Radio Frequency Identification (RFID) tags are characterized by high demand, wide range of applications, and wide geographical (use temperature) span, which puts high demands on the reliability of the product, and it is of great practical significance to study the degradation characteristics at high and low temperatures. Based on Arrhenius theory, this paper proposes a method to quickly calculate the activation energy of electronic tags, to carry out a short time life acceleration test on electronic tags, to predict the life situation through the degradation data of the tag's electric field strength performance, to quickly solve the value of activation energy of the electronic tags, and then to calculate the acceleration factor under different temperature conditions, which provides a theoretical basis for the accelerated test time for electronic tags under different life evaluation standards.

Abstract：In this paper, we designed a quantized convolutional neural network for real-time classification of ECG data, quantized the weights to INT2, applied knowledge distillation to achieve the desired classification results, and deployed it on FPGA. The quantized network after knowledge distillation improved the classification accuracy by 9% over the full precision network. The running results on the FPGA meet the expectations and achieve the required performance to classify four types of ECG signals, left bundle branch conduction block (L), right bundle branch conduction block (R), normal beat (N) and ventricular premature beat syndrome (V), which requires less storage parameter requirements and less resource usage than other quantization methods, and improves the computational speed of the CPU compared to the CPU by 1.5 times, the running time meets the real-time requirement, and is suitable for deployment on small, lightweight wearable devices with limited resources.

Abstract：Aiming at the strict requirements of scanning electron microscope (SEM) high-resolution and all-weather work, a 200 kV high-stability high-voltage DC power supply with high stability and excellent insulation performance was proposed and developed. Experiments show that the system can achieve a switching ripple of 3.42‰ and a power frequency ripple of 2.45‰, and the long-term stability per hour can reach 0.093‰~0.158‰, which can meet the requirements of SEM for ripple and stability.

Abstract：In response to the increased requirements for maintaining balance in power systems and the problem of LED lighting loads in large indoor facilities to serve as non-rotational reserves, we have designed an intelligent lighting system based on BLE Mesh distributed control and the average consensus algorithm. The lighting system utilizes a distributed network constructed using BLE Mesh, combined with the average consensus algorithm, to efficiently control a large number of lighting fixtures, ensuring both power adjustments and user comfort converge to a consistent level. The experimental results demonstrate that all lighting fixtures in the system can achieve the desired power adjustments within a short period, while minimizing the impact on user comfort resulting from power adjustments. The system exhibits advantages such as low network resource consumption, strong stability, and ease of scalability, making it capable of meeting practical requirements.

Abstract：High frequency, high efficiency, and high power density have become important design indicators for AC-DC converters. Traditional quasi resonant flyback converters, due to their reliance on passive resistive components to release leakage energy in the form of thermal energy, do not possess ZVS characteristics, which limits the further improvement of power density. However, active clamp flyback converters can compensate for the shortcomings of quasi resonant flyback converters. This article firstly outlines the working principles of the quasi resonant flyback converter and the active clamp flyback converter, and analyzes the main performance differences between the quasi resonant and active clamp flyback converters. Then, two 48 W quasi resonant flyback converter prototypes and active clamp flyback converter prototypes are designed and manufactured to obtain relevant experimental data as a basis for performance comparison and analysis.