基于批处理和核函数的非线性盲源分离算法  

Nonlinear Blind Source Separation Algorithm Based on Batch and Kernel Function

作　　者：禹华钢[1] 高俊[1] 黄高明[1] 

机构地区：[1]海军工程大学电子工程学院,武汉430033

出　　处：《电讯技术》2011年第10期35-40,共6页Telecommunication Engineering

基　　金：国家高技术研究发展计划(863计划)项目(2010AA7010422)~~

摘　　要：针对基于核函数的非线性盲源分离算法性能对核函数及其参数选择依赖性强这一问题,提出采用批处理方法代替聚类和核主成分分析方法来构造低维近似子空间的正交基,以改进基于核函数的非线性盲源分离算法对核函数及其参数变化的稳健性,并对这种改进的非线性盲源分离算法进行了完整的分析。通过仿真实验,对分离信号与源信号求相似度,可以看到提出的基于批处理的非线性盲源分离算法能够取得更稳健、准确的分离效果。To solve the problem that the performance of nonlinear blind source separation algorithm based on kernel function is dependent on the kernel function and its parameters, this paper proposes using batch methods to construct orthonormal basis for reduced dimension approximate subspace instead of clustering and KPCA metheds. This improved nonlinear blind source separation algorithm based on batch and kernel feature space is investigated firstly, and then is used to improve the robustness to the variety of the kernel function and its parameters. The simulation results illustrate that the algorithm based on batch is more robust and is relatively simple and effective.

关 键 词：信号处理 非线性盲源分离 核函数 聚类 批处理 相似度 

分 类 号：TN911.7[电子电信—通信与信息系统]