The direct current micro-current standard source is widely used in fields such as calibration of electrical instruments,semiconduc-tor testing,and electrostatic detection,involving key technologies like current negative feedback regulation and equipotential shielding.There are mainly three basic implementation methods for the direct current micro-current standard source:resistive,capacitive,and ioniza-tion.Each technology has its own unique advantages and limitations.This paper reviews the research status of domestic and foreign direct current micro-current standard sources in recent years,investigates the current situation of domestic and foreign direct current micro-current standard sources,and points out that China started relatively late in this field and there is a gap compared with the international advanced level.
With the rapid development of the Internet and the widespread deployment of network infrastructure,the number of domain name servers(DNS)has continued to grow,accompanied by their expanding geographical scope.This expansion is characterized by imbalanced regional distribution,complex service hierarchies,and dynamic operational changes,which exacerbate security risks within the DNS ecosys-tem and create challenges for comprehensive oversight.To address these issues,this study establishes a large-scale DNS discovery system and conducts an in-depth analysis of the spatiotemporal evolutionary patterns and hierarchical dependencies among domain name servers.The findings reveal critical insights into the topological evolution of DNS infrastructure,offering actionable support for enhancing cyberse-curity governance and optimizing the strategic deployment of next-generation digital infrastructure.
This work proposes and simulates a high-linearity GaN HEMT device using Technology Computer Aided Design(TC AD).The device features a composite channel based on graded-composite AlGaN and InGaN layers.This composite channel forms a dual-channel dis-tribution consisting of both a three-dimensional electron gas(3DEG)and a two-dimensional electron gas(2DEG).The device achieves a large gate voltage swing(GVS)of 4.1 V,which is 2.2 V higher than that of a conventional abrupt heterojunction HEMT.The peak value of the second derivative of transconductance is 56％lower than that of the conventional device,demonstrating superior suppression of third-order intermodulation distortion(IMD3).Furthermore,the impact of self-heating effects on device linearity is investigated.The proposed de-vice structure holds promise for high-linearity applications.
Secondary radar systems primarily detect cooperative targets through an interrogation-response mechanism and are extensively applied in fields such as air traffic control.To address the issue of declining detection probability for dense tar-gets in traditional secondary radar systems,this pager analyzes the mechanism by which random response delay enhances the detection probability for dense targets,formulates the underlying mathematical problems,and validates the findings through theoretical derivations,calculations,and Monte Carlo simulations.The results demonstrate that the random response delay technique can effectively improve the detection performance of secondary radar systems for dense targets.Furthermore,un-der the premise of meeting detection probability requirements,the number of detectable dense targets exhibits a linear step-like relationship with the number of response positions.
Aiming at the key problems of high missed detection rate of initial fire,significant interference of complex background and diffi-cult balance between edge equipment deployment accuracy and real-time performance in UAV highway slope fire detection,this study pro-poses a lightweight visible light fire detection algorithm based on the improved YOLOv12 framework.The algorithm accurately extracts slope and road areas through instance segmentation technology to eliminate background interference.A small target enhancement feature pyramid(SOPAN)including high-resolution P2 layer and multi-path downsampling mechanism is designed to improve the detection ability of small targets,and a lightweight detection head(LSCD)based on shared convolution is constructed to achieve model complexity com-pression.The experimental results show that on the fusion UAV inspection data set,the proposed model achieves 92.5％mAP detection ac-curacy with only 1.8M parameters,which is 28.3％lower than the benchmark model YOLOv12n parameters.It provides an efficient solu-tion for real-time and high-precision fire warning at the UAV side.
