【摘要】 由时域有限连续信号的部分频段值重构其它频段值的问题，可以转化为求解线性方程Ａ·ｘ＝ｂ的问题。奇异值分解（ＳＶＤ）方法能够清楚地刻画算子Ａ的值域和数值零空间，通过计算信号与所张成值域的正交基的内积，可以预先判断信号重构性能的好坏。数值计算表明ＳＶＤＴｉｋｈｏｎｏｖ正则解比Ｐａｐｏｕｌｉｓ－Ｇｅｒｃｈｂｅｒｇ（ＰＧ）算法在色噪污染情况下有更好的抗干扰性。更多还原

【Abstract】 Signal reconstruction from limited data is completed by finding the minimum norm solution of the linear equation A·x=b . In this paper, we confine ourselves to the time limited signal reconstruction. Because the singular value decomposition (SVD) can describe the ranges and the nullspaces clearly, we can judge the reconstruction result in advance by the inner product of the signal and the orthogonal eigen vectors. Simulations are presented to illustrate the better noise rejection by the SVD method than by the Papoulis Gerchberg algorithm.更多还原