基于连续蚁群算法的Bayesian方位估计快速方法

针对蚁群算法(ACO)在解决高维非线性搜索问题方面的有效性,提出了基于蚁群优化算法的Bayesian最大后验概率方位估计(ACO-Bayesian)快速方法.该方法将Bayesian最大后验概率函数作为蚁群算法的目标函数,选取若干一维高斯函数的加权和作为连续蚁群算法中信息量概率分布函数,经过有限次迭代得到Bayesian方法的非线性全局最优解.仿真结果表明,ACO-Bayesian方法在保持Bayesian方法优良性能的同时,将Bayesian方法的计算量减少到原来的1/14.水池实验结果验证了ACO-Bayesian方法的正确性和有效性,为其工程应用奠定了基础.

Fast Bayesian DOA estimator using continuous ant colony optimization

For the effectiveness of the ant colony optimization algorithm for solving high-dimensional nonlinear search problem, a Bayesian maximum posteriori direction of arrival (DOA) estimation fast algorithm based on the ant colony optimization algorithm (ACO-Bayesian) was proposed. This algorithm adopts Bayesian maximum posteriori probability function as the objective function of the ant colony algorithm, exploits a weighted sum of several one-dimensional Gaussian functions in the sampling process. The global maximum of Bayesian spatial spectrum function can be reached after reasonable iterations. Simulation results show that the proposed algorithm provides similar performance to that achieved by Bayesian estimator, but its computational complexity cost is only 1/14 of original method. The water tank experiment results verified the correctness and validity of the proposed ACO-Bayesian method, which promote them to promising in engineering applications.