The　modal　parameter　identification　procedure　of　civil　engineering　structures　under　operational　conditions　will　need　to　base　itself　on　output-only　measurement　data,　which　presents　a　challenge　that　requires　the　use　of　special　modal　identification　technique.　Stochastic　subspace　identification　(SSI)　algorithm　has　been　developed　in　the　past　few　years.　It　is　a　novel　time　domain　identification　method,　which　directly　uses　operational　response　data　to　identify　the　system　model.　In　this　paper,　a　principal　component　analysis　(PCA)　based　stochastic　subspace　identification　algorithm　is　proposed　where　the　QR　factorization　is　first　applied　to　reduce　the　data　as　the　SSI　does,　while　the　PCA　is　implemented　to　determine　the　system　matrix　A.　The　proposed　SSI-PCA　method　is　demonstrated　using　a　simulated　measurement　data　of　a　frame　structure.　The　obtained　results　have　been　compared　with　those　previously　obtained　from　the　peak　pick　method　in　frequency　domain　and　stochastic　subspace　identification　in　time　domain.　It　is　observed　that　the　PCA　based　SSI　algorithm　is　comparable　in　both　computing　efficiency　and　accuracy.