基于Metropolis Hastings抽样短采样宽带信号方位估计AML算法

针对短采样宽带信号近似最大似然（approximated maximum likelihood,AML）方位估计计算量大的问题，将马尔科夫链〖CD*2〗蒙特卡罗方法与近似最大似然方位估计相结合，提出一种基于Metropolis Hastings抽样的近似最大似然方位估计方法（AMLMH）。该方法将AML算法的空间谱函数作为信号的概率分布函数，并利用Metropolis Hastings抽样方法从该概率分布函数中抽样。研究结果表明，AMLMH方法不但保持了原近似最大似然方位估计方法的优良性能，而且减小了计算量。

AML algorithm for short sampling wideband signal DOA estimation based on Metropolis-Hastings sampling

In view of the approximated maximum likelihood (AML)estimation which is suitable to the short time sampling wideband signals DOA estimation,is non-perfect for its very large computation load,a novel approximated maximum likelihood DOA estimation based on Metropolis-Hasting sampling (AMLMH) is proposed by combining the Markov chain Monte Carlo methods are combined with approximated maximum likelihood DOA estimator.AMLMH regards the power of the AML spectrum function as the target distribution up to a constant proportionality,and uses Metropolis-Hasting sampler to sample from it.Simulations show that AMLMH not only keeps the excellent performance of the original AML but also reduces the computational load greatly.