摘要
针对整周模糊度解算中实时性不强的问题,提出了一种基于实数编码的自适应遗传算法。首先利用卡尔曼滤波求解整周模糊度的浮点解,然后采用排序和多次(逆)双乔里斯基分解对浮点解及其协方差阵进行降相关处理,降低整周模糊度各分量之间的相关性,最后利用已知基线确定模糊度搜索空间,基于实数编码,将自适应遗传算法应用在整周模糊度的搜索解算过程中,求得整周模糊度的最优解。仿真结果表明,本文算法与Lambda算法相比,耗时更少;与简单遗传算法相比,具有更强的收敛能力,是一种高效的整周模糊度快速解算方法。
Aiming at the weak real-time performance in integer ambiguity resolution, an adaptive genetic algorithm based on real-coding is proposed in this paper. The float solution of integer ambiguity is calculat- ed by applying Kalman filter method, the ordering and multi-time (inverse) paired Cholesky decomposition are adopted to decorrelate the float solution and its covariance matrix, Thus, the correlation of each ambi- guity float estimation can be eliminated. With the given baseline, the integer ambiguity search space and the integer ambiguity optimization results can be determined by using adaptive genetic algorithm on the basis of real-coding. The numerical simulation results show that, the integer ambiguities can be resolved by using this propose method in a shorter time than Lambda method, and stronger convergent ability is acquired, which is compared with simple genetic algorithm. It is proven that this algorithm is efficient for fast integer ambiguity resolution.
作者
易清明
易夕冬
石敏
Yi Qingming Yi Xidong Shi Min(School of Information Science and Technology, Jinan University, Guangzhou 510632, China)
出处
《航天控制》
CSCD
北大核心
2017年第3期14-18,共5页
Aerospace Control
基金
广东省科技计划资助项目(2015B090910001)
广州市科技计划资助项目(201604010007)
作者简介
易清明(1965-),女,湖南岳阳人,博士,教授,主要从事卫星导航信号处理的教学与科研工作
易夕冬(1992-),男,湖南岳阳人,硕士研究生,主要研究方向为卫星导航信号处理
石敏(1977-),女,湖北襄樊人,博士,副教授,主要从事多媒体技术和信息安全的教学与科研工作。