This PDF file contains the front matter associated with SPIE Proceedings Volume 13277, including the Title Page, Copyright information, Table of Contents, and Conference Committee information

Conducted research and analysis on existing foundation information systems, and analyzed the requirements and existing shortcomings of future information systems. A space-based intelligent information system based on LEO constellation was proposed, which was designed from three aspects: overall system, constellation orbit, and measurement and control link, and its feasibility was analyzed. A fast collection method for regional information based on wireless sensor networks is proposed for ground terminals. The protocol stack of wireless sensor networks is designed for high bandwidth and low bandwidth regional perception needs, and network integration design is carried out for wireless sensor networks and spacebased systems.

Encrypted traffic classification can effectively supervise and manage the data traffic transmitted in the network. Most deep learning-based encrypted traffic classification models focus on modeling only one characteristic of the network, but the network traffic has both spatial and temporal characteristics, and part of the research adopts the recurrent neural network training method to grasp the temporal characteristics of the traffic, but there is a low efficiency problem when training the model on sequential data. The model is not efficient. In this paper, we propose ASTNet, an encrypted traffic classification model based on spatio-temporal features. Firstly, the session stream is cut according to the length of 784 bytes and then converted into a grayscale map as the input to the spatial feature extraction sub-module of the ASTNet model. Take out the first 8 data packets in the session stream, then intercept 256 bytes in each data packet, sort them according to the timestamp of the data packets, and use the processed time series as the input of the time feature extraction sub-module of the ASTNet model. Then the output features of the two feature modules are feature fused to obtain the spatiotemporal features of the encrypted traffic, and finally the final classification result is output by the classifier. We conducted on the public dataset ISCXVPN2016 (VPN-nonVPN dataset) experiments were conducted to compare the experimental results with the baseline method, and Experimental results show that our model achieves better results in encrypted traffic classification.

Identifying key nodes in communication networks is crucial for network security, fault diagnosis, and strategic deployment of critical nodes. In this paper, we construct a multi-view reliability assessment metric, which comprehensively considers the microscopic, mesoscopic, and macroscopic structures of networks, including end-to-end connectivity, interconnectivity between organizational structures, and overall network connectivity. Moreover, we are inspired by K-shell and propose a method for identifying key nodes based on network organizational structure. By employing the idea of node deletion, it analyzes the variation of network reliability after attacking network nodes, thus achieving key node discovery based on network reliability.

The security of commercial encryption algorithms has always been a focus of attention. The SM4 algorithm, a commercial cryptographic standard, has been recognized by ISO/IEC as an international standard and has shown excellent performance on classical computers. However, with the development of quantum computing, traditional encryption algorithms are facing new challenges. This paper investigates the quantum realization of SM4 algorithm. By applying the Grover algorithm to the SM4 algorithm, we conducted a quantum exhaustive attack and provided an estimation of the required quantum resources. This study offers a novel perspective on the application of quantum computing in breaking commercial encryption algorithms, which holds significant implications for the future development of quantum secure communication and cryptography.

This paper proposes a multi-strategy particle swarm optimization (MPSO) algorithm for optical phased array (OPA) phase noise compensation. The MPSO algorithm combines weight parameters and elite strategies to improve search efficiency and achieve global optimization. Simulation shows that the optimized optical phased array maintains a peak side-lobe ratio (PSLR) greater than 6.699 dB within a range of ±45 degrees. Compared with particle swarm optimization (PSO), genetic algorithm (GA) and stochastic parallel gradient descent (SPGD), the sidelobe suppression effect of the MPSO algorithm is improved by 4.00%, 50.30% and 1.35% respectively. Using a 32-element lithium niobate optical phased array at 1550 nm, the best performance occurs at a deflection angle of 0 degrees with a peak side-lobe ratio of 7.939 dB. The maximum beam scanning angle of 35° is achieved, and the peak side-lobe ratio is greater than 7 dB in the range of -17.5d° to 17.5°. These results demonstrate that the multi-strategy particle swarm optimization algorithm can effectively compensate for phase noise in optical phased arrays.

In this paper, the outage performances of the downlink collaborative non-orthogonal multiple access (NOMA) assisted simultaneous wireless information and power transfer (SWIPT) system are investigated. SWIPT-assisted cooperative relaying transmission offers a sustainable solution to address both performance equity between center and edge users in cells, as well as ensuring fair energy consumption for cell-center users. As far as the concrete is concerned, It is potential to consider the near-end user in NOMA as a repeater in the SWIPT system. The outage probability (OP) of a cooperative NOMA-assisted downlink SWIPT communication system is examined. Specifically, the OP expressions for a two-user NOMA system are derived. The numerical results demonstrate that employing cooperative SWIPT relaying transmission with an optimized power splitting (PS) coefficient significantly enhances the OP for edge users in the cell.

In order to meet the individual communication performance requirements of distributed smart grid operation and operational services, it is crucial to select appropriate communication technologies to build a secure, reliable and economical network. This paper first analyses in detail the performance requirements of new communication business scenarios of distributed smart grid, and reviews the advantages and limitations of current mainstream communication technologies. Then, a comprehensive weighting solution model of network attributes is constructed using subjective and objective evaluation methods. Finally, OPNET simulation software is used to simulate the communication performance of mainstream communication technologies around data throughput capacity, delay, delay jitter, packet loss rate, and the performance of different communication technologies is compared and summarised to provide suggestions on the optimal communication technologies for the new communication services of distributed smart grid.

The development of intelligent warehousing for electric power materials requires an efficient and accurate material identification system. Current research is focused on optimizing RFID technology for tagging and identifying materials in the electric power industry. However, due to the large volume of materials and frequent "multi-tag collisions," existing identification capabilities are struggling to keep up with the rapid development of smart warehousing. To solve this issue, we have created a hybrid neural network-based anti-collision algorithm for RFID, specifically designed for the DFSA anti-collision protocol. The algorithm first predicts the number of tags using a hybrid neural network model. Then, based on the predicted tag count and using tag prefix information, we group and estimate the number of tags in each subset. Finally, we dynamically adjust the frame length for DFSA according to the tag count in each subset, improving the accuracy of RFID technology. Simulation results show that our proposed algorithm outperforms other algorithms in terms of system throughput, the ratio of total time slots consumed to collision time slots, and overall time slot consumption, effectively enhancing RFID identification accuracy.

To satisfy the high data transmission rate requirements of high-speed railway (HSR) wireless communication systems, this paper proposes a reconfigurable intelligent surface (RIS) assisted joint beamforming design algorithm. The algorithm aims to maximize the achievable rate for the HSR communication system while complying with per-antenna power constraints (PAPC) at the base station (BS) and uni-modular constraints on the reflection phase shifts at the RIS. Considering the nonconvex nature of the optimization problem, an alternating optimization (AO) algorithm is employed, decomposing it into two sub-problems: at the BS, an accelerated subgradient (AS) algorithm based on gaussian distribution is proposed to determine the closed-form optimal solution of the transmit covariance matrix; at the RIS, an accelerated minorizationmaximization (AMM) algorithm based on square iteration method is proposed to approximate the optimal solution of the reflection phase shift matrix. Simulation results show that the proposed joint beamforming algorithm effectively addresses the beamforming design challenges in large-scale multiple-input multiple-output (MIMO) HSR communication systems, significantly enhancing the system's achievable rate.

Aiming at the problem of large error of long-term prediction model based on ITU-R P.533 model, this paper proposes a prediction algorithm of long-term and short-term memory neural network based on deep learning technology. This algorithm uses ionospheric data obtained by vertical sounding technology to construct ionospheric model and simulate short-wave propagation path, so as to predict the future medium and short-term maximum usable frequency (MUF) of short-wave communication. Simulation results show that the root mean square error (RMSE) and mean absolute error (MAE) of the predicted maximum usable frequency (MUF) are 1.3666 and 1.0891 for the next 24h.The RMSE and MAE of the MUF for the next 48h are 1.3723 and 1.0771 respectively, and the RMSE and MAE of the MUF for the next 72h are 1.5707 and 1.2519, respectively. LSTM shows good validity and accuracy in ionospheric MUF short-term prediction, and provides valuable reference for ionospheric research and short-wave communication.

The main purpose of System-of-systems simulation modeling is to measure the contribution of equipment to the overall effectiveness of the System-of-systems. Space-Earth Integrated Communication System-of-systems exhibits significant characters of large scale, large number, various types of communication units and complex environment. In order to meet the industrial development of Space-Earth Integrated Communication, it is urgent to carry out research on System-of-systems Modeling, Simulation, and index evaluation. It brings great challenges such as simulation confidence requirement, extraordinary contradiction between simulation precision and efficiency, and high requirements of generalization level on simulation model. In this work, the simulation methods, simulation models, simulation technologies and simulation applications are intensively studied for the System-of-systems combat for Space-Earth Integrated Communication. This paper proposes the model construction ideas, thoroughly investigates the three dimensions of model construction in System-of-systems: Elements, correlations, and operation environments in System-of-systems, and forms a set of modeling framework of System-of-systems simulation model. This model can provide a foundation for the simulation and deduction of Satellite Communication System-of-system. And a multi-level Simulation model architecture is proposed base on interhierarchy modeling methodology, which broke through the key technologies of universal code framework design, optimization of satellite communication full link simulation efficiency, decoupled optimization of simulation flow for satellite resource allocation under different systems. The simulation indexes and results presents the characteristic of high-dimension, high-redundancy, and high-relationship, and a resonable reduction of indexes and efficient neural network model for effectiveness evaluation are introduced. With the comparison between simulated data and measured data, the validity of the model is verified which provides a solid support for the communication System-of-systems simulation. The research results provide an integrated solution of System-of-systems simulation platform, and an application of SpaceEarth Integrated Communication System-of-systems Modeling and Simulation is presented and verifies the research results. The modeling idea and inter-hierarchy simulation methods embodied in this dissertation can also be extended to other satellite System-of-systems application.

This paper designs a K-band two-dimensional beam scanning liquid crystal reflective array based on sub-array division. The phase shifting unit adopts a "工"-shaped dipole structure, and a new sub-array division arrangement is adopted, by dividing the 24×24 liquid crystal reflective array into three parts, namely, the smallest phase shifting unit, 2×2 sub-array unit, and 3×3 sub-array unit, reducing the number of feedlines and achieving two-dimensional beam scanning. Simulation results show that the 24×24 reflective array designed in this paper, with the new sub-array division arrangement in the Kband, achieves a ±40° beam scanning in the E plane and H plane, and the array gain can maintain above 20dBi, with good directionality and side lobe levels below -5dB.

In modern wireless communications and radar systems, accurate direction finding technology is crucial for signal processing and positioning applications. This paper deeply explores the impact of the number of array elements and aperture size on direction finding performance, aiming to provide a theoretical basis for antenna array design. By establishing antenna array models of different uniform circular arrays, their direction finding accuracy is analyzed. Finally, simulation experiments are performed to verify the correctness of the theoretical analysis.

The accurate recognition of parameters in the photovoltaic (PV) model is crucial for the assessment, adjustment, and tracking of the maximum power point in the PV system, which has significant theoretical and practical value. This paper proposes an adaptive differential evolution with an accelerated exploitation mechanism (AESHADE) for estimating parameter in PV models. In AESHADE, the algorithm's parameters fully utilize the historical evolutionary information of successful individuals and make adaptive adjustments. During the exploitation phase, when the algorithm stagnates, parameters are adjusted to accelerate convergence. To validate the efficacy of AESHADE, it is applied to the single diode model (SDM), double diode model (DDM), triple diode model (TDM), and Photowatt-PWP201 module model (PhotowattPWP201 MM). The experimental results demonstrate that, compared to seven others optimization algorithms, the AESHADE algorithm exhibits superior problem-solving precision on PV models.

This article mainly discusses the application of non-invasive load identification technology in electrical equipment monitoring. Through the research on key technologies such as load data collection, event detection, load feature extraction, and electrical equipment identification, accurate monitoring and analysis of the operating status of electrical equipment have been achieved. This technology has the advantages of non invasiveness, low cost, and strong real-time performance, providing important support for energy management and the development of smart grids.

A lightweight indoor mobile robot based on ROS is proposed to address the problems of traditional AGV robots mostly relying on patrolling operations, which are costly and limited by fixed paths. The Raspberry Pi 4B was chosen as the master device, the STM32 microcontroller as the slave device, and the combination of hardware such as LiDAR to carry the mobile robot. By introducing two SLAM map building algorithms, Gmapping and Cartographer, and optimizing the parameters of Cartographer algorithm, we get the high quality maps. Aiming at the problems of high computational complexity and inefficiency of traditional RRT* algorithm in path planning, the idea of partitioning is proposed. The range of primary and secondary regions is obtained through calculation, the number of regional nodes is limited, the weight of primary regions is improved to avoid excessive exploration in secondary regions, and finally simulation is performed in MATLAB2022a. The experimental results show that the improved RRT* algorithm reduces the path length by more than 7.6% and the search time by more than 47.8% for both different maps. Therefore it can be utilized in production lines to improve productivity.

Avionics Full Duplex Switched Ethernet (AFDX) technology, AFDX bus has become one of the most advanced aviation buses in the field of airborne networks due to its high real-time, high reliability, and low latency, and is currently widely used in advanced civil aviation passenger aircraft such as A380 and B787 [1]. The function of the AFDX network-side system-on-chip is to ensure the safe and reliable transmission of data between avionics systems, and it is also widely used in civil airliners, so it is important to verify it. However, due to its complex design, traditional chip verification methods cannot find chip problems in time, so it is particularly important to use virtual simulation technology to verify it.

In order to realize the accurate recognition of bridge cable surface damage, an image recognition method based on improved YOLOV8 model is proposed. The image information of the bridge cable surface is collected by the rope climbing robot, which is an intelligent detection device for bridge cables, and then the image data set with defects on the cable surface is selected. Firstly, Diverse Branch Block (DBB) was introduced to construct a C2fDBB module to replace the C2f module in the original algorithm and enhance the capability of multi-scale feature extraction. Secondly, in the original YOLOV8 model, the MSCA attention module was combined with the common CBAM attention module, and the MCASAM attention module was innovatively designed and added to the last layer of the feature extraction network, which improved the multi-scale feature capture ability of the model and the detection sensitivity of small targets. The experimental results in the homemade cable defect dataset show that the average accuracy of the improved algorithm mAP50 is 0.9357 and the average detection time is 0.176 compared with the original algorithm, which are 6.97% and 6.82%, respectively. The experimental results show that the improved YOLOv8 model has better performance in defect detection.

The next-generation Internet plays a significant role in carrying 5G, cloud computing, the Internet of Things, and other important technologies. It serves as a crucial network infrastructure for the development of digital China. However, during the initial design of the Internet, the security threats posed by untrustworthy activities were not adequately considered. Next-generation Internet equipment only forwards messages based on the IPv6 destination address, and this leaves the system vulnerable to attacks that involve forging the source address. Attackers use this method to impersonate the victim and launch distributed denial of service (DDoS) attacks. This type of attack is particularly damaging as it is cost-effective, difficult to trace, does not require a large number of "zombie hosts," and is highly aggressive, making it challenging to defend against. This poses significant harm to the next-generation Internet. SAVNET (Source Address Validation in Intra Domain Networks and Inter-domain Networks) offers an effective solution to the problem of Internet source address forgery. This technology employs a hierarchical trust federation and encryption label mechanism to efficiently verify source addresses, thus enhancing the security and credibility of the network. The paper introduces the application of SAVNET in typical scenarios of Internet Service Provider (ISP) networks, along with related source address verification technology. It also analyzes the technical principles, implementation methods, and challenges associated with this technology.

In this paper, we designed a two-layer emergency communication network integrating remote backhaul and on-site coverage to address the issues of complex terrain and limited public network coverage, which makes emergency rescue communication unable to be guaranteed. On the one hand, this network can support on-site coverage of mobile nodes to meet the needs of on-site command and scheduling, on the other hand, it supports long-distance backhaul, which can connect the site with the rear. On site testing has shown that the system can meet the joint command needs of multiple departments in emergency situations and achieve the need for command extension to the site.

Tropospheric scatter as a promising communication for over-the-horizon between remote geographic areas, has regained its predominance in military wireless communication recently when considering cable systems are not feasible. However, since the propagation path loss is relatively high, the troposcatter (short for tropospheric scatter) system must rely on highpower transceivers. Diversity reception is the most effective method to smooth channel deep fading and suppress large signal fluctuation. In this paper, the multipath diversity reception architecture model of troposcatter is established at first, which aims to analyze the diversity receiving performance. Focusing on maximal ratio combing (MRC), we derive the exact expression of the average bit error performance of binary phase shift keying (BPSK) and quadrature phase shift keying (QPSK) under single-channel reception or diversity reception in the flat or selective fading channels. More importantly, we investigate and simulate the anti-fading performance of diversity receiving based on MRC over fading channel. Numerical and simulation results demonstrate diversity promotes the receiving performance of anti-multipath fast fading. When diversity multiplicity increases, the average bit error rate (BER) of the system will decrease continuously, to improve the reliability of the troposcatter communication.

Detecting data flow anomalies within Wireless Sensor Networks (WSNs) and reaching the server can promptly identify faults and issue maintenance alerts, thereby ensuring safe and reliable operation of the system. However existing methods can utilize single variables in data streams to perform most anomaly detection tasks, they often ignore the correlation between variables, resulting in reduced detection performance. Therefore, our approach constructs a generative adversarial network model based on Graph Attention Network (GAT) and employs a dual autoencoding structure in both the generator and discriminator to learn the latent representation of the entire time series. The results of ablation experiments and comparison experiments on two real datasets with baseline models show that GAT-GAN outperforms the former in detecting data anomalies

Conventional short-term load forecasting methods for main network, which are mainly based on load change feature extraction, are affected by the complexity of forecasting and cannot adapt to multiple operation scenarios, which affects the accuracy of the final forecast. Therefore, a short-term load forecasting method for main grid considering multiple operation scenarios is designed. The uncertainty probability density of main network short-term load prediction is determined, and according to the distribution characteristics of load, a suitable probability density function is selected for fitting to ensure the accuracy of subsequent load prediction. Based on multi-scenario operation, the model in this paper is constructed, and different operation scenarios are defined according to historical load data, weather conditions, economic activities and other factors, so as to arrive at more accurate load forecasting values. The short-term load forecasting model is optimized in parallel, and the distributed computing framework is used to process the load data in parallel to meet the forecasting performance requirements.

Energy trading strategies are essential for Multi- microgrid system (MMS) because they lower microgrid costs and lessen the erratic effects of microgrid systems on the distribution grid. This paper proposes a multi-buyer-multi-seller system allowing free choice of multiple trading objects based on non-cooperative game strategies. Each microgrid aims to minimize its own transaction costs and influence the trading behavior of multiple microgrids through price signals. This allows each microgrid to recover as much as possible of its excess or deficient energy through inter-grid trading. In the process of solving the game model, to simplify the solution process, the power and tariff in the strategy space are optimized by the buyer's microgrid and the seller's microgrid, respectively. To ensure fairness between the microgrids, separate optimization of the microgrids is conducted in each round of the game following a random order. During the optimization of the buyer's microgrid power, the purchase order of power from the seller's microgrid is also randomly determined. Finally, the effectiveness of the proposed strategy is verified by four interconnected microgrids.

The safety of the distribution network is of utmost importance. Although the integration of distributed generation can affect relay protection strategies, optimization can fully leverage its advantages. Distributed generation alters the power flow distribution in the network, impacting the settings of protection devices during short-circuit faults. Therefore, relay protection strategies need optimization. This paper proposes two solutions: first, analyzing from the perspective of relay protection strategies, adjusting the settings and operation modes of protection devices; second, optimizing the protection devices themselves by configuring more reliable equipment. Simulation validates the effectiveness of the optimized strategy, demonstrating successful fault isolation without expanding the outage area. The optimized protection strategy significantly enhances the safety and reliability of the distribution network.

Traditional security detection methods are difficult to meet the real-time, accurate, energy-saving, and scalability requirements of wireless sensor networks. Therefore, design a multi-layer security detection scheme for NAWL-ILSTM and apply it to wireless sensor networks. The concept of mobile Agent is introduced, estimate node spacing using signal attenuation characteristics. The wireless sensor network is divided into multi-layer structures, and the static state of network nodes and the transfer state during the transmission task are analyzed, combined with the LSTM network and NAWL optimization algorithm of online updating mechanism. The introduction of real-time observation data and on-line updating of parameters realize multi-layer security detection of wireless sensor networks. The experiment shows that the false alarm rate of the design method is as low as 2%, and the recall rate is as high as 0.91, and the average detection time is generally low, especially when the number of iterations is 30, 40 and 50, the detection time is the shortest, which is 0.41s, 0.37s and 0.38s respectively.

The current power communication system in China is mainly based on ground communication technologies such as power optical fiber, industrial Ethernet, 4G, 5G wireless public networks, etc. After extreme weather and natural disasters, the power communication system is easy to be damaged, resulting in the inability of the power system. Power space-based IoT business integration system can collect video, image, and power service data and carry out AI calculation. It uses satellite communication means to achieve all-day, all-weather, stable data transmission. The communication channels are safe and reliable. The system can ensures reliable communication on various levels and types of power communication platforms, and ensures stable operation of the power grid.

To address the issues of low efficiency and accuracy in the generation of switching operation tickets, a method for generating switching operation sequences based on Generative Adversarial Networks (GAN) is proposed. Firstly, the SDAR (Sequence Decomposition And Reconstruction) module is used to perform multi-scale decomposition and reconstruction of the original operation task sequences, capturing both the overall and local features of the operation task sequences. Secondly, an A-LSTM (Attention-Long Short Term Memory) network is introduced to enhance the accuracy of the generated switching operation sequences. Finally, the generator of the proposed model is constructed based on the SDAR module and the A-LSTM network. Experimental results show that the proposed model achieves a 47.10% improvement in RMSE and a 4.26% improvement in R² compared to LSTM, and a 27.33% improvement in RMSE and a 1.73% improvement in R² compared to WGAN.

The safe operation of transmission lines is crucial for the transmission of electrical energy, and arc sag and safety distance have attracted extensive research as important parameters for measuring whether transmission lines are operated safely or not. In this paper, we present an innovative edge computation model for monitoring arc sag and safety distance in transmission lines. The model combines real-time data collection with a localized computational process to ensure timely and accurate detection of risks faced by transmission lines. By utilizing the distributed capabilities of edge computing, the model reduces latency and improves monitoring response time.

Due to the characteristic of low-power, the bluetooth mesh network has a great potential in distributed smart grids due to their low-powers. However, due to duplicate forwarding brought on by data broadcasting, traditional routing protocols based on flooding algorithms increase energy consumption. It is obviously unsustainable for battery-powered mesh networks relying on battery power that handlea large amount of data. Therefore, this study designs a novel routing protocol for multi-node, multi-level tree network topology for the specific needs of battery-powered device-dependent nodes in distributed smart grids for multiple application scenarios such as environmental monitoring. The routing protocol is designed by abstracting and analyzing the wireless self-organizing network topology in different scenarios and optimizing the design based on this topological abstraction. Simulation results show that the proposed protocol reduces the energy consumption of sensor nodes by 52.8% compared to the traditional Bluetooth Mesh, and improves the communication reliability at the same time. This improvement provides a new technical path for the wide deployment of Bluetooth Mesh networks in battery-powered smart grid applications.

Aiming at the problems of performance fluctuation and unstable training of edge-end devices during training, this paper proposes an adaptive federated learning device selection strategy based on edge-end performance prediction. The method aims to solve the problem that the traditional federated device selection strategy does not consider the performance of edge-end devices, which leads to a large amount of wasted arithmetic and communication resources and slow convergence speed. Specifically, the method establishes a linear regression model based on the overhead of each resource by collecting the historical performance data of edge-end devices to realize the performance prediction based on the amount of local data; and realizes an efficient and stable federation training process in dynamic edge-end environments by realizing a device selection decision module based on the dobby slot machine algorithm, which adaptively learns the linkage between the performance characteristics of the devices and the device selection decision. Experimental results show that the method proposed in this paper can select stable devices for updating as much as possible, and has significant advantages in ensuring high global convergence speed and high training accuracy

The 3 d optimization model of traditional overhead transmission conductor is difficult in data acquisition and update, limited model accuracy, computational complexity and high resource consumption, insufficient consideration of external factors and lack of real-time monitoring and feedback, based on this, the 3 d optimization model of overhead transmission conductor based on UAV aerial survey modeling is proposed. Based on the aerial triangulation method, control the work of the information acquisition device, collect the position information of the device, use the Canny operator edge detection algorithm to detect the edge information in the image; use tilt photography 3 D reconstruction technology to provide the 3 D scene model of overhead transmission wire. The experimental results show that the error rate of the proposed method is 0.93%, and the error rate is small. The predicted values of the iterations of 120 and 140 times are 0.7dB and 0.3dB larger, respectively, which is consistent with the measured results and provides better support for the optimization, operation and maintenance of the transmission system.

The influence of urban terrain on radio transmission is relatively obvious. How to solve the problem of radio transmission loss in urban terrain is of great significance for improving the communication quality and optimizing the communication network layout. Through the analysis of ray tracking algorithm theory, put forward for the optimization of the algorithm, using the twoway mirror method and particle swarm algorithm improvement optimization, comparative analysis of simulation results, PSO in relatively simple three-dimensional model of two-way mirror method can achieve more than 70% of the optimization effect, achieve the expected research.

Node positioning is one of the key technologies in wireless sensor networks, and positioning error and positioning coverage are two key indicators to measure the performance of node positioning, which have always been the focus of the research community. The article focuses on six representative mobile anchor node path planning positioning algorithms, including Σ-Scan, LH, LMAT, M-curves, N-curves, and TGS. Through comparative analysis, the effectiveness of these algorithms in reducing positioning error and improving positioning coverage is deeply discussed. To comprehensively and objectively evaluate the performance of the algorithms, this paper introduces the least squares method as the positioning algorithm to ensure the accuracy of the evaluation results. The experimental results show that all six algorithms demonstrate sensitivity to changes in the number of nodes and communication radius in terms of positioning error and positioning coverage.

The Network Data Analytics Function (NWDAF) is an integral component of an operator’s communication network, responsible for the automated sensing and analysis of network data. It plays a crucial role throughout the entire lifecycle of network planning, construction, operation, optimization, and management. This study represents a continuation and deepening of previous research efforts, utilizing variants of the Transformer model for time series forecasting. The experimental results underscore the significant research value of the currently popular Transformer and its variant models. It is important to note, however, that the performance of these models can vary depending on the specific application scenarios and data characteristics.

In order to solve the uncertainty of renewable energy output and load demand in the integrated energy microgrid, this paper proposes a source-load coordination optimization scheduling method based on hybrid two-stage robust integrated energy microgrid (IEM). Firstly, the framework of the transaction system in the integrated energy microgrid was established. Secondly, in order to cope with the challenge of source-load multiple uncertainties in the integrated energy microgrid, a hybrid two-stage robust optimization method based on multi-scenario probability was proposed on the basis of the traditional two-stage robust optimization, and the nested Column-and-Constraint Generation (C&CG) algorithm was used to decompose the hybrid two-stage robust optimization model into the main problem and the sub-problem for cyclic solving. Finally, the effectiveness of the model and the solution algorithm is verified by a case study, and the economy and ability to resist the fluctuation of uncertain factors are analyzed.

A bi-level distributed optimization for networked micro-grids with weak interconnection is presented. In the upper-level, the scheduling model of the networked micro-grids is optimized by the alternating direction method of multipliers (ADMM) to reduce the operation cost. The lower-level internal optimization model of micro-grid based on cooperative game is solved by particle swarm optimization (PSO) to minimize the cost. This optimization approach can enhance the economy and reliability of networked micro-grids, achieving effective cooperative control and management.

Based on GaAs HBT technology, a narrowband high linearity power amplifier chip operating in the 2.48-2.51 GHz frequency range was designed. The amplifier adopts a three-stage amplification structure to ensure high gain and high output power. Designed is a temperature-compensated adaptive bias circuit structure, which effectively suppresses the selfheating effect of HBTs, ensuring high linearity of the power amplifier. The output matching circuit was optimized to suppress harmonic components. Test results demonstrate that within the 2.48-2.51 GHz frequency range at room temperature, the output saturated power is 37.5 dBm with 1 dB compression, achieving an efficiency of 41%. The Error Vector Magnitude (EVM) is less than 4%, Adjacent Channel Power Ratio (ACPR1) is less than -30 dBc, and Second Adjacent Channel Power Ratio (ACPR2) is less than -40 dBc. The power amplifier utilizes an unpackaged die combined with external circuit integration, where the input and output matching circuits are achieved through off-chip substrate matching, reducing the chip area and facilitating later-stage debugging, meeting the application requirements of mobile communication systems.

Amidst the growing diversification of user-side energy demands, accurate multi-load forecasting has emerged as a critical factor for effective planning and optimization of microgrids. As versatile distributed energy systems, microgrids necessitate adept management and dispatch of various loads - spanning electricity, heat, cooling, and energy storage. Nevertheless, these loads exhibit intricate spatiotemporal correlations and heterogeneities, posing formidable challenges for load forecasting. Current methods predominantly concentrate on time series correlations, neglecting the spatiotemporal data's heterogeneity, thereby inadequately serving practical applications. To tackle these challenges, this paper introduces a novel multi-load prediction model for microgrids, rooted in the Spatial-Temporal Synchronous Graph Convolutional Network (STSGCN). This model integrates Graph Convolutional Network (GCN) and Long Short-Term Memory (LSTM), synergistically capturing both the correlations and heterogeneities within spatiotemporal data. Simulation outcomes underscore the model's effectiveness, with the STSGCN model demonstrating a significant reduction in average absolute error - approximately 13% compared to traditional LSTM, 7% versus GCN, and 5% GCN-LSTM prediction models. This underscores the model's potential to revolutionize multi-load forecasting in microgrids.

In the context of economic expansion and elevated living standards, the imperative for power grids to maintain real-time balance and operational security has become increasingly critical. This research investigates the complexities and dynamics of medium- to long-term load forecasting within a multi-industry framework, highlighting the criticality of accurate forecasting for the strategic planning and sustainable evolution of power grids. The study underscores the importance of integrating industry-specific user datasets to enhance the precision of predictive models, thereby facilitating more informed decision-making in power grid development and the efficient integration of new energy resources.

The new generation of information technology, represented by artificial intelligence, 5G, big data, etc., has widely empowered all walks of life. The "carbon peak" and "carbon neutrality" of power are promoting the green transformation and upgrading of the industry, and are promoting the development of the power grid to a smarter, more ubiquitous and more friendly energy Internet. The power network is complex, the facilities and equipment are complex, the number of customer groups is huge, and the value of power big data has not been fully released, so it is urgent to introduce digital means to promote business transformation. The digital power business is in the ascendant, and it is urgent to rely on the resource advantages of the main business of the power grid to expand and strengthen the industrialization of digital business, open up new blue oceans, and cultivate new kinetic energy. However, at present, artificial intelligence has not been effectively practiced in many business scenarios of power enterprises, and its role in business operation, management, and service has not been significantly highlighted. At the same time, it is also seen that the proportion of artificial intelligence applications in many fields such as intelligent substations, intelligent production safety management and control, and data empowerment is generally low. Therefore, it is necessary to explore artificial intelligence to help the digital transformation of Liaoning power grid, which can not only effectively accelerate the transformation and development of power grid construction and ensure the safety of power operation, but also comprehensively promote the rapid development and progress of the power field.

This article explores the application of deep learning (DL) algorithms in power system load forecasting. With the continuous advancement of the construction of new power systems, traditional load forecasting models designed based on time series analysis are no longer in line with the actual needs of the current power system. To improve the accuracy of load forecasting, this paper attempts to transform time series data into images for analysis, and then use deep learning methods widely used in the field of image processing for power load forecasting. Convolutional neural network (CNN), as a commonly used image processing algorithm, has been applied in time series data processing. However, data is still processed as a sequence matrix, and the advantages of CNN algorithm in processing image matrices have not been fully utilized. Therefore, this article proposes a CNN algorithm based on sequence to image conversion (STI-CNN). Firstly, the load sequence is converted into a load image. Then, a dual branch deep network model is used to accurately cluster the input data. Finally, the STI-CNN model is used for load prediction. Matlab simulation experiments show that the proposed STI-CNN model has excellent performance in different prediction metrics and has higher prediction accuracy.

In the context of the current era where energy demands are increasingly growing, accurate electric load forecasting is crucial for ensuring the stability and economic efficiency of power systems. This paper presents a novel electric load forecasting method based on tax and electricity joint data, known as the DQPSO-LSTM model. This model integrates Quantum Particle Swarm Optimization (QPSO) with Long Short-Term Memory networks (LSTM), aiming to enhance forecasting accuracy. To overcome the limitation of QPSO's tendency to converge prematurely, we introduce a dynamic nonlinear varying inertia weight strategy, which enhances the algorithm's global search efficiency and local search capability. By balancing extensive exploration in the early stages and rapid convergence in the later stages, the model parameters are optimized. Validation using electricity consumption data and tax data from Heilongjiang Province from 2018 to 2021 demonstrates that the DQPSO-LSTM model significantly outperforms benchmark models such as LSTM, ARIMA, and Prophet in terms of prediction error (RMSE), mean absolute percentage error (MAPE), and the coefficient of determination (R²), highlighting the significant advantages of the proposed method in electric load forecasting.

With the increasing complexity of electromagnetic environment in the modern battlefield, the multi-beam direction finding system requires higher and higher channel consistency of RF receiving front-end. It is necessary to carry out real-time amplitude correction or phase correction through self-checking power division network. In this paper, a self-checking power distribution network is proposed, and its reliability is improved by optimizing the switch design, and the correction accuracy is improved by introducing the difference between the working branch and the self-checking branch. The self-checking power division network can realize fast, efficient and real-time correction. And through correction, the amplitude consistency is ≤±0.5dB.