压缩感知的例子。重建是用在稀疏域上求最小零范数的方法。首先把零范数凸松弛为一范数，然后变成线性规划问题，从而求得最优解。(An example of the compressive sensing. Its reconstruction is based on L0-norm minimization.)