面向5G-A低慢小目标检测的曲线拟合拆峰技术OA

As a pillar industry of the low-altitude economy,unmanned aerial vehiclespresent detection challenges due to their small size,low flight altitude,slow speed,and susceptibility to being obscured by complex urban environments.To effectively detect Low,Slow,and Small(LSS)unmanned aerial vehicles targets,the radar echo characteristics of 5G-A low-altitude scenarios is analyzedand a detection technique based on curve fitting and peak separationis proposed.The Doppler spectrum is extracted from the range cell under test,and the number of peaks and their locations on the spectrum curve are obtained using a smoothed derivative peak-finding method.The Cell-Averaging Constant False Alarm Rate(CA-CFAR)threshold is used to eliminate noise peaks,retaining only valid peaks.The peak location,peak height,and peak width of the valid peaks are derived through curve fitting.The presence or absence of the target is detected by determining whether there are any valid peaks with non-zero frequency,and the target's range and velocity are obtained accordingly.Field test results demonstrate that the proposed algorithm has better detection performance compared to the traditional CA-CFAR and CA-CFAR combined with Moving Target Indicator(MTI).