项目名称: 运动宽带目标的波达方向估计研究
项目编号: No.61302168
项目类型: 青年科学基金项目
立项/批准年度: 2014
项目学科: 无线电电子学、电信技术
项目作者: 冯大航
作者单位: 中国科学院声学研究所
项目金额: 26万元
中文摘要: 运动宽带目标波达方向(DOA)估计是目标跟踪领域的一个重要问题。本项目旨在通过对传统宽带DOA估计方法和粒子滤波算法的研究,建立新的运动宽带目标的DOA估计模型,引入角度函数(角度随时间变化的函数),利用相邻快拍之间的相关性,寻找角度函数参数的最优化求解方法。同时由于宽带目标运动,会产生多普勒效应,以及运动过程中发动机转速的变化等因素,目标的频谱特征也会随之发生变化,即目标的频谱也是时间的函数。在新模型的框架下,进一步引入频谱函数,实现对角度函数和频谱函数参数的同时估计,以及利用两者之间的相关性提高角度估计的精度和算法的鲁棒性,同时利用频谱函数的估计结果还可以为多运动目标的DOA估计提供解决途径。
中文关键词: 波达方向估计;贝叶斯;宽带运动目标;粒子滤波;目标跟踪
英文摘要: Direction of arrival (DOA) estimation for moving wideband target is a key problem in target tracking. In this project, a novel DOA estimation model for moving wideband target will be studied based on the conventional DOA estimation methods of wideband signal and the technique of particle filter. In the new model, the DOA function (the DOA of target varies with the time) is introduced, and the parameters of the DOA function will be estimated optimally through the correlation of the adjacent snapshots. The spectrum of the moving target also varies with the time due to many reasons, such as the effect of Doppler, the changing of engine speed. In the new model, the spectrum of moving target is further introduced, and the parameters of DOA function and the spectrum function are estimated simultaneously. The correlation of the DOA function and the spectrum function can be used to improve the accuracy of DOA estimation and robustness of DOA estimation. The estimation of the spectrum can also be used to solve the problem of DOA estimation of multiple moving targets.
英文关键词: Direction of arrival;Bayesian;wide-band moving target;particle filter;target trajectory