The invention discloses an OFDM (Orthogonal Frequency Division Multiplexing) system parametric channel estimation and equalization method based on compressed sensing technology. The channel estimation method comprises the following steps: 1) an OFDM frequency domain at a pilot subcarrier receives data, and parameter estimation is carried out on the channel based on a channel parametric model: firstly, according to a MDL (Minimum Description Length) criterion, a multipath number estimation value L<^> of the channel is obtained; then, the ESPRIT (estimation of signal parameters by rotational invariance techniques) technology is used for estimating delay parameters for each path, and multipath delay estimation values tau<^>=[tau<^>1, tau<^>2,..., tau<^> < L<^>>] are obtained; and 2) a parameter estimation result is used for building a redundant dictionary matrix, and a channel response estimation value is solved through a BPDN algorithm. Based on parametric channel modeling, the signal parameter estimation technology is combined with a compressed sensing frame, the channel estimation precision and the communication efficiency are improved, and the communication bit error rate is reduced. Compared with the traditional channel estimation method, the method of the invention has obvious performance advantages.