In industrial networks, more and more intelligent applications put forward new requirements for deterministic guarantee ability of end-to-end transmission. Most of current research works on deterministic technologies focus on guarantee of their own network certainty, but ignore end-to-end determinism requiring cooperation of multiple deterministic technologies. Based on application scenario of nuclear power industry, a wide-area end-to-end deterministic network architecture with Flexible Ethernet (FlexE) and Time-Sensitive Networking (TSN) fusion is proposed, and an architecture is elaborated from two aspects of control plane function and data plane cooperative scheduling. In real wide area network environment, based on the dual-motor collaboration and industrial machine vision application scenario, the performance of the architecture is tested and verified. Results show that the architecture can meet the needs of applications for wide-area end-to-end deterministic networks, and offer good application value.

Driven by the ever-increasing supply of computational resources, the parameter size of Large Language Models(LLMs) continues to expand and their task performance in natural language processing has become more superior. However, there are still limitations when faced with reasoning problems, especially in common-sense reasoning or mathematical problems. Chain of Thought(CoT) significantly improves its ability to solve problems in different domains by guiding the model to generate reasoning steps. In this paper, we not only sort out the theoretical foundation system and technical evolution of CoT from the perspective of training method, but also further discuss application scenarios such as government service and enterprise digitalisation. Finally, in the light of the development trend of Artificial Intelligence (AI), the paper discusses the essential role of CoT in the development of LLMs towards a higher cognitive level from the perspective of the degree of AI, and points out the challenges and technical bottlenecks that need to be solved at the present time.

Based on the principle of Orthogonal Time Frequency and Space (OTFS) modulation, this paper designs an OTFS waveform scheme based on Zero Suffix (ZP) protection. Methods of synchronization, channel estimation, and the detection algorithm based on delay-time domain Maximum Ratio Combining (MRC) are presented, and the MRC detection algorithm is simplified. Hardware implementation schemes of channel interpolation and the MRC detector are given. And the Field Programmable Gate Array(FPGA)hardware implementation of the proposed OTFS system waveform is carried out to verify the feasibility of the key algorithms of the designed OTFS system. Test results show that the designed OTFS system has good performance in resisting doubly selective fading.

In this paper, a Wireless Powered Communication Network (WPCN) based on Active Simultaneously Transmitting and Reflecting Reconfigurable Intelligent Surface (ASTAR-RIS) is proposed. The communication network is mainly composed of four parts: Power Station (PS), Sensor Node Groups (SNGs), ASTAR-RIS and Access Point (AP). The operation process of the communication system is mainly divided into two stages: Wireless Energy Transfer (WET) stage and Wireless Information Transfer (WIT) stage. Energy Splitting (ES) mode is adopted in the wireless energy transmission phase, and Time Switching (TS) mode is adopted in the wireless information transmission phase. This paper aims to optimize the phase shift parameters and communication resource allocation of ASTAR-RIS to maximize system throughput. Since the optimization problem is non-convex, this paper uses an alternate optimization algorithm to solve the problem. Firstly, the problem is divided into two parts according to the coupled variables. The optimal solutions of the variables in these two parts are solved by Semidefinite Relaxation (SDR) and Fractional Programming (FP) respectively. Experimental results show that the communication scheme proposed in this paper can provide higher performance gain for the system.

The Internet of Things (IoT), as one core area of 6G development, plays a crucial role in driving network architecture changes and supporting core application scenarios. However, IoT systems suffer from energy imbalances and short network lifecycles, which severely restrict the improvement of data collection efficiency. With the rise of Unmanned Aerial Vehicle (UAV) technology, its high maneuverability can effectively construct Line of Sight (LOS) communication links, thereby improving communication speed. This has great application value in data collection of IoT systems and can solve the problem of low data collection efficiency caused by the short lifecycle of IoT networks. To this end, UAVs are used to collect data from ground IoT devices and build a data collection and transmission link for air-to-ground collaboration. An intelligent data collection method based on Deep Reinforcement Learning (DRL) is proposed. In addition, a predictive neural network is designed to further improve data collection efficiency by predicting network data at the Base Station (BS) side, thereby achieving the goal of reducing IoT device energy consumption and extending network lifespan. Simulation results show that the proposed data collection algorithm has good performance advantages in terms of device energy consumption and energy balance, and is superior to traditional data collection algorithms. At the same time, the proposed data collection network architecture can extend the network lifespan by 1.2 times when the predicted data accounts for 12.5%. In addition, simulations have shown that the designed predictive neural network outperforms other compared networks in terms of Mean Squared Error (MSE) and Mean Absolute Error (MAE) metrics.

Compared with a different-color backgrounds, recognizing and detecting cucumber fruits under uniform-color backgrounds remains a key challenge due to limited distinguishing features and increased susceptibility to occlusion and background interference. To address this, we propose YOLO-ACG, a detection network based on YOLOv11n. An Adaptive Dynamic Downsample (A-Down) module is introduced, combining deformable convolution and channel attention to achieve adaptive cross-scale feature sampling. A Ghost_HGNetV2 architecture is designed, where the High-resolution Group Stem (HGStem) reduces input channels to extract efficient intrinsic features, and the Ghost_HGBlock applies knowledge distillation to enhance feature representation. A Context and Spatial Feature Calibration Network (CSFCN) network structure is introduced, which includes Context Feature Calibration (CFC) and Spatial Feature Calibration (SFC). The CFC module aggregates context information relevant to each pixel, while the SFC module leverages calibrated spatial features to ensure accurate understanding of spatial layout the image. Together, they enable the network to more precisely distinguish cucumber fruits from backgrounds with similar colors. Experimental results show that the improved model achieves 4.64 percentage points increase in precision, recall by 5.07 percentage points, F1 by 4.89 percentage points, and mAP by 4.48 percentage points. Ablation and comparative experiments confirm that YOLO-ACG significantly reduces false positives and missed detections, offering effective technical support for cucumber fruits recognition in complex, uniform-color environments.

In recent years, Transformer-based visual models (e. g. , Swin Transformer) show good prospects in visual tasks, however, these methods usually focus on reducing signal distortion between original and reconstructed data, while ignoring perceptual quality. Considering that the conventional Mean Square Error (MSE) loss fails to reflect perceptual and semantic quality effectively, we propose a weighted loss function combining MSE and Learned Perceptual Image Patch Similarity (LPIPS), and accordingly construct a Swin Transformer-based semantic communication framework, called Swin Transformer with LPIPS-based Joint Source-Channel Coding (STL-JSCC) method, which significantly enhances image reconstruction quality and semantic consistency. For performance evaluation, two semantic-aware metrics are introduced: the Images Semantic Deviation (ISD) value and Iamges Semantic Similarity(ISS). These indicators form a joint perceptual-semantic evaluation system, which breaks through the limitations of traditional evaluation methods. Experimental results show that the proposed STL-JSCC outperforms other models in all the indexes, verifying the significant potential and advantages of the proposed method in improving the image reconstruction quality and semantic extraction capability.

In smart grids, the presence of numerous high-power electrical devices and communication sensing equipment severely hinders signal transmission in positioning systems. To address the challenge of accurately locating weak signals in a Reconfigurable Intelligent Surface (RIS)-assisted Non-Orthogonal Multiple Access (NOMA) system under interference from multiple base stations and communication users, this paper considers the impact of multiple small base stations and multiple users in a smart grid environment. A horizontal positioning error of the target user is used as the evaluation metric. While ensuring the Quality of Service (QoS) for communication users, the proposed method jointly optimizes base station power, multi-user interference, and power allocation factors. The Lagrangian dual method and sub-gradient approach are employed to solve the non-convex optimization problem caused by multiple users and small base stations. Simulation results demonstrate that, under the same resource allocation, the proposed RISNOMA integrated sensing and communication system significantly reduces the average positioning error compared to traditional NOMA-based system.

Predicting trajectories of key individuals plays an important role in preventing potential criminal activities, optimizing emergency response, and intelligence analysis. Application of this technology by public security departments helps maintain social stability, improve urban management efficiency, and improve economic development. However, existing techniques face challenges in adapting to dynamic environments, neglecting the scope of social influence, and influence quantification of neighborhood moving objects. A novel model for predicting long-term trajectory areas of key individuals based on destination-intention learning by integrating spatio-temporal queries, is proposed. Firstly, aiming to solve the problem of capturing the spatio-temporal features of moving object trajectories, a key individuals trajectory prediction model called Spatio-Temporal Multiple Attention (STMA) is introduced. It can enhance the model sensitivity to the change of behavioral features by capturing temporal dependencies and spatial interactions through temporal and spatial attention modules, respectively. Secondly, in order to cope with the problem of quantifying the social influence, a social force function is constructed to simulate the social influence of pedestrians. The virtual contour construction method and the social force function can accurately simulate dynamic behaviors and improve the efficiency of influence capture. Experiments based on real-world traffic datasets show that, compared to the state-of-the-art trajectory prediction algorithms, STMA demonstrates higher accuracy and reliability in long-term and short-term trajectory prediction. In terms of long-term forecasting, the STMA model achieves an average accuracy rate of 54.3%, outperforming Sophie by 29.3%, Social Spatio Temporal Graph Convolutional Neural Network (S-STGCNN) by 13.4%, Conditional Generative Neural System (CGNS) by 36.8%.

In future 6G Internet of Things (IoT) systems, extensive deployment of pivotal technologies such as high-frequency millimeter waves and terahertz spectrum makes it possible for wireless transmission among network devices situated in the near-field region. As a byproduct, Dynamic Metasurface Antenna (DMA) and other small-sized antenna arrays have been widely applied in this scenario due to their advantages in transmission efficiency, physical size, and power consumption. And related research has received increasing attention. Aiming to improve the energy performance of receivers in near-field wireless transmission, a downlink near-field wireless Simultaneous Wireless Information and Power Transfer (SWIPT) system based on DMA is proposed. Under the condition of satisfying the minimum transmission rate requirements of all information users, an efficient solution for jointly optimizing the tunable frequency response matrix of DMA and the digital precoding vector is proposed for this optimization problem. In addition, the influences of factors such as the distance between users and the minimum Signal to Interference plus Noise Ratio (SINR) on the system performance are also discussed on this basis. Simulation results show that the scheme proposed in this paper can effectively improve the joint performance of wireless information and power transmission compared with other existing technologies.