利用隔离小生境混合蛙跳方法联合估计波达方向和多普勒频率

发布时间：2018-08-25 07:42

【摘要】：为了高效准确地联合估计波达方向(DOA)和多普勒频率,提出了基于二维MUSIC谱和隔离小生境混合蛙跳算法的联合谱估计方法。首先,利用状态空间模型构造包含DOA和多普勒频率信息的广义可观测矩阵。然后,利用MUSIC算法和广义可观测矩阵构造联合估计的二维谱函数,将参数估计问题转化为多峰函数优化问题。最后,引入混合蛙跳算法(SFLA)并利用隔离小生境技术对混合蛙跳算法进行改进,使其适应于二维MUSIC谱多个极值的优化搜索,降低搜索计算量,得到DOA和多普勒频率的估计值。实验结果表明:在考虑信噪比为0dB的两个独立信号源的情况下,利用本文方法估计得到的DOA和多普勒频率的均方根误差分别为0.23°和0.005 5rad。该方法具有控制参数少、计算量小、参数自动配对等特点。
[Abstract]:In order to estimate direction of arrival (DOA) and Doppler frequency accurately and efficiently, a joint spectrum estimation method based on two-dimensional MUSIC spectrum and isolated niche hybrid leapfrog algorithm is proposed. Firstly, a generalized observable matrix containing DOA and Doppler frequency information is constructed by using the state space model. Then, the MUSIC algorithm and the generalized observable matrix are used to construct the two-dimensional spectral function of the joint estimation, and the parameter estimation problem is transformed into the multi-peak function optimization problem. Finally, the hybrid leapfrog algorithm (SFLA) is introduced and improved by using isolated niche technique, which is adapted to the optimal search of multiple extreme values of two-dimensional MUSIC spectrum, reduces the amount of searching computation, and obtains the estimated value of DOA and Doppler frequency. The experimental results show that the RMS errors of DOA and Doppler frequencies estimated by the proposed method are 0.23 掳and 0.005 rads, respectively, when two independent signal sources with a signal-to-noise ratio of 0dB are considered. The method has the advantages of less control parameters, less computation and automatic matching of parameters.
【作者单位】：吉林大学通信工程学院;中国人民解放军61785部队;
【基金】：国家自然科学基金资助项目(No.61201368) 吉林省产业技术研究与开发项目(No.JF2012C013-3)
【分类号】：TN911.23

【相似文献】