The　sparse　nature　of　positioning　in　the　spatial　domain　allows　the　use　of　compressed　sensing　theory　for　wireless　positioning.　Compressed　sensing-based　positioning　algorithms　can　reduce　the　number　of　online　measurements　to　a　great　degree　and　achieve　high　positioning　accuracy　at　the　same　time,　which　makes　compressed　sensing-based　positioning　algorithms　extremely　attractive　for　tunnel　positioning.　However,　traditional　localization　methods　based　on　compressed　sensing　are　mostly　ranging　and　unsuitable　for　the　energy-constrained　low-loss　wireless　sensor　network.　Therefore,　a　coal　mine　tunnel　personnel　positioning　algorithm　based　on　non-ranging　compressed　sensing　is　proposed　in　this　article.　According　to　the　connectivity　information　between　the　target　nodes　and　the　sensing　nodes,　the　algorithm　designs　a　non-ranging　compressed　sensing　positioning　model　and　establishes　a　database　for　the　positioning　area,　which　provides　a　solution　to　the　problems　of　low　positioning　accuracy　and　time　delay.　Experiment　and　simulation　results　show　that　the　proposed　algorithm　can　achieve　higher　positioning　accuracy　and　better　robustness.　©　2021　Acta　Press.　All　rights　reserved.