摘要
针对超定情况下盲源分离算法及其在雷达信号分选中的应用进行了研究。在分析了超定盲源分离自然梯度算法最终不能稳定收敛的原因基础上,构造了一种适合于源信号数动态变化的自适应盲分离算法。新算法不仅克服了已有算法最终不能稳定收敛的缺点,极大地简化了算法的计算量与复杂度,而且在源信号数随机减少或增加时仍能够达到较好的分离效果。仿真结果验证了算法的收敛稳定性与分离有效性。
The over-determined blind source separation algorithm and its application in radar signal sorting are studied. Based on analysis of the reason why the over-determined blind source separation of natural gradient algorithm can＇ t ultimately converge stably, an adaptive blind separation algorithm suitable for dynamically changing source number is constructed. The new algorithm not only over- comes the shortcoming of the existing algorithm such as eventually not implementing stable convergence,so it greatly simplifies the com- putational complexity of the algorithm, and it can obtain better separation effect when the number of source signals decreased or increa- ses. Simulation results prove the convergence of the algorithm and its effectiveness for blind source separation.
出处
《无线电工程》
2014年第1期20-23,共4页
Radio Engineering
基金
国家自然科学基金资助项目(61174207)
关键词
源信号数
动态变化
自适应
盲源分离
number of source signals
dynamic changes
self-adaptation
blind source separation
