中文摘要：

在无线传感器网络（ WSN）中，传统的处理方式是采用奈奎斯特技术对信号进行采样并重构，而随着信号频率的增加，应用奈奎斯特技术会使成本急剧增加，这是人们所不乐见的。针对这一问题，近年来出现一种新的技术即压缩感知技术，它能利用更少的数据和合适的重构方法得到更精确的原始信号。将稀疏贝叶斯学习（ SBL）和压缩感知联合起来，形成了一种在噪声的情况下更好重建可压缩信号的方法，并进一步将这种方法应用在WSN中，可以在误差允许的范围内有效控制测量数据的维数，所以在保证了一定的误差的同时还减少了成本，提高了算法的效率。

英文摘要：

In wireless sensor networks,signal is sampled and reconstructed using the technology of Nyquist in the past. But it requires a substantial increase in the cost with the growth of the signal frequency,which is that people do not like to see. Recently a new technology is emerged,which is called compressive sensing technology,is a good way to solve this problem. Compressive sensing can use less data and appropriate reconstruction method to get a more accurate original signal. Put Sparse Bayesian Learning （ SBL） and compressive sens-ing together to form a better reconstruction compressible signal under the noise. This method can effectively control the dimension of measurement data within the range of allowed error in WSN,so you can ensure a certain degree of error while reducing the cost,impro-ving the efficiency of the algorithm.