In　this　paper,　a　Data　De-noising　Clustering　(DDC)　algorithm　for　data　aggregation　is　proposed.　By　considering　the　noisy　data　of　sensor　nodes,　the　algorithm　introduces　a　Weighted　Moving　Average　(WMA)　and　improves　it,　then　apply　it　to　data　de-nosing.　Moreover,　we　utilize　the　spatial　correlation　between　nodes　to　divide　nodes　into　different　clusters　and　present　a　correlation　degree　for　cluster　head　selection　so　that　the　data　of　the　cluster　head　have　a　low　distortion　on　their　correlated　data　while　the　cluster　head　is　used　to　represent　the　cluster　and　send　data　to　the　sink.　The　experiments　result　show　that　the　resulting　achieved　by　DDC　can　provide　sensing　data　with　higher　accuracy　and　less　energy　consumption　compared　with　other　algorithms.　©　2015　Taylor　&　Francis　Group.