Location　information　is　very　crucial　for　the　sensing　data　in　wireless　sensor　networks　(WSN).　But　localization　precision　and　cost　are　two　challenging　problems　for　practical　localization　applications.　We　proposed　a　novel　algorithm　to　tackle　these　challenges.　Our　collaborative　sparse-anchored　scheme,　CSA,　utilizes　multi-hop　anchors　collaboration　to　improve　localization　accuracy　and　ratio,　especially　for　unknown　nodes　on　edge　of　network.　Furthermore,　it　overcomes　ill-equation　problem　of　minimal　mean　square　algorithm　(MMS).　Error　analysis　demonstrates　that　position　error　and　ranging　error　of　some　anchors　can　incur　huge　error.　So　it　is　necessary　to　judge　the　feasibility　of　reference　anchor.　Our　experimental　results　verify　validity　and　accuracy.　It　improves　feasibility　and　cost　of　WSN　positioning　techniques,　significantly.　©　2008　IEEE.