文献类型：会议论文

英文题名：A Novel Blind Source Separation Approach Based on Invasive Weed Optimization

作者：Li, Zhu-cheng[1,2];Huang, Xiang-lin[1]

第一作者：Li, Zhu-cheng;李著成

通讯作者：Huang, XL[1]

机构：[1]Commun Univ China, Fac Sci & Technol, Beijing 100024, Peoples R China;[2]Beijing Union Univ, Business Coll, Beijing 100025, Peoples R China

第一机构：Commun Univ China, Fac Sci & Technol, Beijing 100024, Peoples R China

通讯机构：[1]corresponding author), Commun Univ China, Fac Sci & Technol, Beijing 100024, Peoples R China.

会议论文集：International Conference on Communication, Network and Artificial Intelligence (CNAI)

会议日期：APR 22-23, 2018

会议地点：Beijing, PEOPLES R CHINA

语种：英文

外文关键词：BSS; IWO; Optimization algorithm; Negentropy

摘要：Traditional optimization algorithms for Blind Source Separation (BSS) are mainly based on the gradient, which requires the objective function to be continuous and differentiable, and have some defects such as slow convergence speed or poor accuracy of the solution. To solve these problems, a novel BSS approach based on Invasive Weed Optimization (IWO) is proposed in this paper. By maximizing a negentropy-based objective function, simulation experiments confirm the effectiveness of the proposed algorithm.