基于最大似然估计的空间谱测向技术

介绍了基于最大似然估计（简称ML）的空间谱测向,ML是一种高分辨算法,常用于空间测向和谱估计等方面。文中首先介绍了ML测向原理,由于ML是一个非线性、多维的极值问题,需要全局极值的多维搜索,其计算量非常的巨大,因此采用了AP算法进行ML算法的优化,最后通过均匀圆阵对非相干信号和相干信号进行了ML测向的计算机仿真,仿真结果表明：一个阵列对空间多个信号的分辨能力除了与信噪比有关外,还与入射信号是否相干以及两相干信号的相位差有关,ML算法可以直接对相干的入射信号进行测向,但是信号的相干性会影响ML算法测向的角度分辨率,在两信号相关的情况下,两信号间的相位差越大则测向精度越高。

Direction Finding of Spatial Spectrum Based on Maximum Likelihood Estimation

It is introduced to direction finding of spatial spectrum based on maximum likelihood estimation, ML is a kind of high resolute arithmetic which is applied to the direction finding of spatial spectrum and spectrum estimate. The principle of ML direction finding is introduced in the paper firstly, ML needs multidimensional search and its calculated quantity is very high, so ML is optimized by AP arithmetic. At last, the simulation of MI. direction finding is operated for non coher- ent and coherent signal by even circle array, it is showed by simulated conclusion that the resolution of direction finding is re lated to not only SNR but coherence of signals, moreover, which is related to phase difference of coherent signal, direction finding of coherent signal can be carried out by ML arithmetic, but the resolution of direction finding can be influenced by co- herence of signals, while two signals are coherent, the larger phase difference is, the higher precision of direction finding is.