With　the　development　of　5G　technology,　the　Internet　of　Everything　is　becoming　possible,　and　the　communication　traffic　of　mobile　terminals　will　explode.　Meanwhile,　the　requirements　of　mobile　terminals　in　terms　of　fluency　and　the　computing　power　of　applications　are　also　becoming　more　stringent.　However,　the　computing　power　of　mobile　devices　is　always　limited　due　to　their　portability.　Edge　computing　can　offer　a　timely　manner　by　offloading　tasks　of　the　mobile　device　to　nearby　cloudlet.　Therefore,　the　computing　tasks　can　be　processed　quickly　nearby　network　edge,　which　can　effectively　reduce　the　system　delay.　Although　there　are　many　researches　on　cloudlet　placement　technology,　how　to　optimize　the　cloudlet　placement　in　a　given　network　to　improve　the　performance　of　mobile　applications　is　still　an　open　issue.　This　paper　mainly　proposes　a　particle　swarm　optimization　algorithm　based　on　genetic　algorithm　(PSO-GA)　to　optimize　the　cloudlet　placement　in　a　wireless　metropolitan　area　network,　aiming　at　reducing　the　average　response　time　for　users　to　process　tasks.　The　simulation　results　show　that　the　PSO-GA　approach　performs　better　in　user　service　quality　and　reduces　system　average　response　time　compared　with　other　cloudlet　placement　schemes.　©　2019,　Springer　Nature　Singapore　Pte　Ltd.