基于FastICA的低信噪比雷达信号分选算法

针对传统的基于参数的信号分选系统已无法适应当前复杂情况下的雷达信号分选问题，将基于独立分量分析(ICA)的盲源分离算法引入雷达信号分选算法.快速ICA(FastICA)算法结合了定点迭代和非高斯最大化算法，具有稳定性好、收敛速度快、计算量小等优点.但该算法对噪声非常敏感，无法在低信噪比情况下进行信号分选.针对这一缺点，引入同步累加平均降噪算法，并结合信号均衡、平滑处理进行改进，使得新算法在低信噪比情况下对雷达信号进行分选.仿真表明改进后的算法在低信噪比情况下具有良好的分选效果，并保留了原算法的优点.

Low Signal to Noise Ratio Radar Signal Sorting Algorithm Based on FastICA

The traditional parameter-based signal sorting system cannot adapt to the problem of radar signal sorting in the current complex situation， and the blind source separation algorithm based on independent component analysis(ICA)is introduced into the radar signal sorting algorithm. FastICA algorithm combines fixed-point iteration and non-Gaussian maximization algorithm. It has the advantages of good stability， fast convergence and small calculation. However， this algorithm is very sensitive to noise and cannot be performed with low SNR. Aiming at this shortcoming， a synchronous cumulative average noise reduction algorithm is introduced， and signal equalization and smoothing to improve the original algorithm are combined， so that the new algorithm can sort the radar signals with low SNR. Simulation results show that the improved algorithm can achieve good sorting effect under low SNR and retain the advantages of the original algorithm.