基于分形维数的自适应窗口DAVAR改进算法  

Improved DAVAR Algorithm for Adaptive Window Based on Fractal Dimension

作　　者：王梓帅 刘星 闫曌 杨银芳 喻佩 WANG Zishuai;LIU Xing;YAN Zhao;YANG Yinfang;YU Pei(School of Electronic Information Engineering,Xi’an Technological University,Xi’an 710021,China;Northwest Industrial Group Co.,Ltd.,Xi’an 710043,China;Xi’an North Electro-optic Scientific and Technology Co.,Ltd.,Xi’an 710043,China)

机构地区：[1]西安工业大学电子信息工程学院,西安710021 [2]西北工业集团有限公司,西安710043 [3]西安北方光电科技防务有限公司,西安710043

出　　处：《弹箭与制导学报》2021年第5期129-132,共4页Journal of Projectiles,Rockets,Missiles and Guidance

摘　　要：针对动态Allan方差(DAVAR)法在分析弹载环境下MEMS陀螺输出信号时存在功率泄露和噪声辨识精度低等缺陷,提出一种基于分形维数自适应窗口的改进DAVAR算法。利用分形维数值表征随机信号突变特征,实时改变窗长截取信号,并将结果绘制在三维图形中。仿真实验表明:该方法能够实时跟踪MEMS陀螺输出的突变,并准确评估动态误差中各噪声的变化特征。Aiming at the shortcomings of dynamic Allan variance(DAVAR) method in analyzing the output signal of MEMS gyroscope in a missile-borne environment, such as power leakage and low noise identification accuracy, an improved DAVAR algorithm based on adaptive window of fractal dimension is proposed. The fractal dimension value is used to characterize the sudden change characteristics of random signals, the window length is changed in real time to intercept the signal, and the result is finally drawn in a threedimensional graph. The simulation experiment shows that the method can track the sudden change of the output of the gyro in real time, and accurately evaluate the change characteristics of the noise in the dynamic error.

关 键 词：MEMS陀螺 动态Allan方差 分形维数 自适应滑动窗 

分 类 号：TJ765[兵器科学与技术—武器系统与运用工程]