论文部分内容阅读
用极大似然估计 (MLE)得到到达信号的方向 (DOA) ,在统计性能方面要比其它一些理论优越 ,但是由于该方法为一种多维参数估计 ,采用常规搜索方法 ,精度受到网格限制 ,不能任意逼近最优解 ,并且容易收敛到局部最优。而遗传算法是一种有导向的随机搜索方法 ,它具有适用条件宽松 ,有较大的概率收敛到全局最优等优点。在此通过改进的遗传算法 (IGA) ,较好地解决了一般搜索算法存在的不足 ,计算机模拟实验证明其可行。
The direction of arrival (DOA) of the signal is obtained using Maximum Likelihood Estimation (MLE), which is superior to other theories in statistical performance. However, since the method is a multi-dimensional parameter estimation, the conventional search method is used and the accuracy is limited by the grid , We can not approach the optimal solution arbitrarily, and it is easy to converge to the local optimum. The genetic algorithm is a directed random search method, which has the advantages of loose conditions, greater probability of convergence to the global optimum. In this paper, through the improved genetic algorithm (IGA), to better solve the shortcomings of the general search algorithm, computer simulation shows that it is feasible.