基于WSNM的气象雷达风力涡轮机杂波抑制算法OA
Wind Turbine Clutter Suppression Algorithm for Meteorological Radar Based on WSNM
风力涡轮机杂波(WTC)严重干扰气象雷达对灾害性天气的监测.针对基于核范数最小化(NNM)的矩阵补全(MC)方法因气象信号观测矩阵奇异值缺乏锐截止性、难以精确逼近秩函数而导致恢复精度不足的问题,提出一种采用加权 Schatten-p 范数的非凸矩阵补全算法.首先,对受 WTC 污染的距离单元进行填零处理,在多普勒频域构造低秩 Toeplitz 矩阵,形成零元素随机分布的观测矩阵.其次,利用加权 Schatten-p 范数逼近秩函数,构建非凸矩阵补全模型.最后,基于增广拉格朗日乘子法(ALM)与交替方向乘子法(ADMM)框架求解模型,实现气象信号的高精度恢复及 WTC 抑制.在实测气象雷达数据上的实验结果表明,与 NNM 模型相比,加权 Schatten-p 范数最小化(WSNM)模型能更有效抑制杂波,显著提升气象信号的恢复精度.
Wind Turbine Clutter(WTC)severely interferes with meteorological radar monitoring of hazardous weather.To address the problem that the Matrix Completion(MC)method based on Nuclear Norm Minimization(NNM)suffers from insufficient recovery accuracy due to the lack of sharp cutoff in the singular values of the weather signal observation matrix,which makes it difficult to accurately approximate the rank function,a nonconvex matrix completion algorithm using the weighted Schatten-p norm is proposed.First,the range cells contaminated by WTC are zero-filled,and a low-rank Toeplitz matrix is constructed in the Doppler frequency domain,resulting in an observation matrix with randomly distributed zero entries.Second,the weighted Schatten-p norm is employed to approximate the rank function,and a nonconvex matrix completion model is established.Finally,the model is solved using the framework of the Augmented Lagrangian Multiplier(ALM)method and the Alternating Direction Method of Multipliers(ADMM),achieving high-precision recovery of weather signals and effective WTC suppression.Experimental results on real meteorological radar data demonstrate that,compared with the NNM model,the Weighted Schatten-p Norm Minimization(WSNM)model can suppress clutter more effectively and significantly improve the recovery accuracy of weather signals.
邱存银;蒋意扬
江苏电子信息职业学院,江苏 淮安 223003南京科瑞达电子装备有限责任公司,江苏 南京 211106
信息技术与安全科学
风力涡轮机杂波低秩Toeplitz矩阵矩阵补全加权Schatten-p范数交替方向乘子法
wind turbine clutterlow-rank Toeplitz matrixmatrix completionweighted Schatten-p normalternating direction method of multipliers
《现代信息科技》 2026 (18)
24-29,6
评论