基于盲源分离算法的水下无线传感网信号重建

Signal Reconstruction of Underwater Wireless Sensor Networks Based on Blind Source Separation

部署在水下的无线传感网,工作环境极其复杂,各种海洋噪声会污染信号并影响传感器的位置。为了从含噪的混合信号中准确有效地分离出未知信号,提出一种有定位预处理的含噪盲源分离算法。利用少数配有GPS接收器的信标节点对未知传感器节点进行精确定位和小波去噪法去除混合信号中的强高斯白噪声,以提高盲源分离算法的性能。再利用盲源分离算法对预处理过的混合信号进行分离。仿真结果表明,采用三边定位法对未知传感器节点定位后,虽然对实时性略有影响,但分离效果明显优于没有定位的情况。

The working environment of the underwater wireless sensor network is very complex. All kinds of ocean noises can pollute the signals and affect the position of the sensors. In order to separate the unknown signals accurate- ly and effectively from the noisy mixed signal, a blind source separation algorithm is proposed： In order to improve the performance of blind source separation algorithm, we first used the beacon nodes equipped with GPS receivers to pinpoint the location of unknown sensor nodes and used wavelet denoising method to remove strong Gauss noise of the mixed signal. Then a blind source separation algorithm was used to separate the mixed signals that have been prepro- cessed. The experimental simulation results show that the separation result relocated using the three side location method is much better than the result un - relocated, although it has a slight influence on the real - time perform- ance.