Compressed sensing is a new theory of collection and processing data,in which a sparse signal can be reconstructed successfully from far fewer data/measurements.The quality of Channel estimation which is seemed as one of the OFDM key techniques impacts on overall system performance directly.In this paper,we consider channel estimation using the compressed sensing for the sparse channels such as ultra-wideband channel and underwater acoustic channel in OFDM systems.This work leads to a common describe of sparse channel estimation based on compressed sensing.Other work is focused on expatiations for the compressed theory in several points and summarizing the methods of signals collection and processing based on compressed sensing. Theoretical analysis and numerical simulations show that the compressed sensing method can serve as an advantage in reducing sampling rate and improving channel estimation precision.