Ang Li

Superconducting nanowire single photon detectors (SNSPDs) exhibit excellent performance in the near-infrared band, but their application range is limited by the detection efficiency and detection speed. In this paper, we propose a high-efficient and ultralow-filling-factor design scheme based on Fano resonance which is excited by a one-dimensional silicon dielectric grating. This design can reduce the kinetic inductance of the nanowires while maintaining the active sensing area of SNSPDs, which is beneficial for improving the recovery time of the detector. Meanwhile, the absorption efficiency of the nanowires can be enhanced by the Fano resonance. Taking λ = 1550 nm as an example, numerical simulations are performed using commercial simulation software based on the finite-difference time-domain method. Under idealized structural conditions, the absorption efficiency of the Niobium Nitride (NbN) superconducting nanowires can exceed 98% when the filling factor of the nanowire is only 10%.

Smart transportation systems use VANETs for real‐time vehicle‐roadside infrastructure communication. These channels enable collision prevention, traffic synchronization, and cooperative driving. Network congestion, signal interference, hardware failures, and malicious assaults may create communication abnormalities in VANET systems due to their open and dynamic nature. The complex and quickly changing patterns of vehicle communication networks make static threshold or statistical anomaly detection methods insufficient. This paper presents a real‐time VANET communication channel monitoring system using AI‐IAD to overcome these constraints. AI‐IAD's hybrid AI architecture models dynamic vehicular communication behavior via adaptive feature fusion and edge‐assisted inference, making it innovative. The system uses multi‐metric learning to identify modest network behavior changes by monitoring packet delivery ratio, communication delay, signal intensity, and channel use. Unlike centralized detection systems, AI‐IAD uses edge‐based processing at roadside units (RSUs) to identify anomalies quickly and reduce network communication. The model also uses traffic‐aware adaptive learning to adapt to vehicle density, mobility patterns, and channel conditions in urban and highway situations. The Vehicular Reference Misbehavior (VeReMi) dataset, a benchmark dataset produced from actual traffic traces and DSRC communication logs, is used to test the proposed AI‐IAD platform. Testing in simulation‐based VANET scenarios shows that the AI‐IAD framework outperforms existing anomaly detection methods with 94.6% detection accuracy, 95.0% precision, and 94.2% recall. The framework's 3.9% false positive rate and 28 ms detection latency provide quick anomaly identification for safety‐critical applications. The system has 92.5% robustness and scalability in large vehicle networks, achieving steady convergence after 500 training epochs. These findings show that the AI‐IAD architecture considerably improves VANET communication channel reliability and security, enhancing intelligent transportation system resilience.

Object detection in UAV aerial imagery plays a pivotal role across a wide spectrum of applications. However, existing detection models continue to face significant challenges stemming from small object scales, dense spatial distributions, and highly complex backgrounds. To address these challenges, this paper proposes a novel dual-backbone network model named WCDB-YOLO. The core innovation of this work lies in introducing a “target-context decoupled perception” paradigm, which utilizes two structurally complementary backbone networks to separately process local object features and global background information: one backbone focuses on extracting fine-grained local features of objects, while the other innovatively incorporates a wavelet convolution module to efficiently model the global contextual semantics of complex scenes with minimal computational cost by constructing a large receptive field. To further enhance the scale adaptability for small objects, a Dilation-wise Residual (DWR) module is designed, which employs parallel convolutional branches with different dilation rates to achieve dynamic adaptation to multi-scale small object features. Additionally, the model optimizes the feature pyramid structure by integrating high-resolution P2/4 features into the detection head, significantly improving the localization accuracy of tiny objects. Experimental results on the VisDrone dataset show that the proposed method achieves an 8.4% improvement in mAP50 over the baseline YOLOv11s model and outperforms current state-of-the-art (SOTA) approaches. This work presents a highly accurate and robust solution for small object detection from UAV platforms in complex environments.

The demand for energy in factories is huge. With abundant energy supply, factories can raise their production capacity to a new height. Energy is the material basis of domestic economic development. As the largest developing country in the world, energy shortage has become an urgent problem for China. In this paper, an Internet of Things based on big data ecosystem is proposed to analyze the energy consumption of the factory and build a model. The Internet of Things technology of big data ecosystem can be summarized as a technology that uses information sensing devices to complete the transaction and network connection according to the protocol content. A total of 853,000 power distribution operations were carried out in the power grid. In 2019, the average ratio of decision tree algorithm, machine learning algorithm, and machine learning algorithm was 36.8%, 37.4%, and 43.5%, respectively. Compared with the three methods, the method in this paper increased by 37.9% year-on-year and reduced the power outage by 2.63 million households, which is equivalent to a corresponding reduction of 35 users per operation. The functional requirements of the IoT energy consumption analysis system in factories based on big data ecosystem are reflected in three aspects: energy consumption monitoring and management, power control and management, and energy consumption supervision and analysis. Based on the management of energy consumption monitoring and power control through the software platform, the functional requirements of the system are analyzed.