The objective of this paper is to develop on-line system parameter estimation and damage detection technique from the response measurements through using both the Stochastic Subspace identification (SSI) and Recursive Covariance-Driven Stochastic Subspace identification (RSSI-COV) approaches. To avoid time-consumption of SVD in RSSI the Extended Instrumental Variable version (EIV-PAST) is used in SSI-COV. From numerical study the reliable control parameters of SSI and RSSI methods are examined. To reduce the effect of noise on the results of identification, discussion on the pre-processing of data using recursive singular spectrum analysis technique (rSSA) is also presented to remove the noise contaminant measurements, so as to enhance the stability of data analysis. Results from using both Data-drive and Covariance-driven SSI methods are discussed. Finally, through experimental study, the recursive rSSA-SSI-COV method is applied to identify the system dynamic characteristics with time-varying model parameters.