摘要
边缘似然最大化算法在远场信号源定位的应用已有研究,但是目前仍然没有将其推广延伸至近场信号。由于近场信号源与远场信号源之间存在着延伸关系,将上述理论自远场源延伸至近场源的研究不仅具有充分的可行性,也具有很大的实用价值。该方法将在边缘似然最大化理论远场源情况参数估计的基础上,避开传统估计方法的缺点,一方面设计并实现合理的算法流程,对算法的性能进行统计分析,另一方面对其做更进一步的延伸和研究探索,设计出在近场源情况下的全新参数估计算法,实现边缘似然最大化理论自远场源至近场源应用的推广。已设计的算法虽然在源数不多的情况下取得了良好的效果,但是在源个数增多的情况下性能有所下降。所提方法提高了模型的稳定性,并通过实验证明了该改进的有效性,具有理论意义与实际应用价值。希望其成果能够对声呐、移动通信和地震勘探中的目标定位和跟踪等技术的发展起到积极的促进作用。
The marginal likelihood maximization algorithm has been applied in far⁃field source localization,but it has not yet been extended to near⁃field source at present.Due to the extended relationship between near⁃field source and far⁃field source,the study of extending the above theory from far⁃field source to near⁃field source not only has feasibility,but also has great practical value.On the basis of far⁃field source parameter estimation based on the marginal likelihood maximization theory,this method get rid of the shortcomings that occur to the traditional estimation methods.On the one hand,a reasonable algorithm flow is designed and realized,and the algorithm performance is statistically analyzed.On the other hand,it is further extended and explored,and a new parameter estimation algorithm in the case of near⁃field source is designed,so as to realize the popularization of marginal likelihood maximization theory from far⁃field source application to near⁃field source application.Although the existed algorithms achieve good results when the number of sources is small,their performance decreases when the number of sources is increased.The proposed method improves the stability of the model and proves its effectiveness by experiments,so it has theoretical significance and practical application value.This achievements can play a positive role in promoting the development of target positioning and tracking technologies in sonar,mobile communication and seismic exploration.
作者
郑超
左炜亮
ZHENG Chao;ZUO Weiliang(General Manager Office,Yan’an Jiude Safety Emergency Technology Co.,Ltd.,Yan’an 716099,China;College of Artificial Intelligence,Xi’an Jiaotong University,Xi’an 710049,China)
出处
《现代电子技术》
2023年第23期28-35,共8页
Modern Electronics Technique
基金
国家自然科学基金青年项目(62201447)。
关键词
边缘似然估计
信号源定位
远场源
近场源
参数估计
目标跟踪
marginal likelihood estimation
signal source localization
far⁃field source
near⁃field source
parameter estimation
target tracking
