To　solve　the　problem　that　the　second-order　fuzzy　function　can　not　deal　with　related　noise,　as　well　as　the　problem　of　large　computation　based　on　fourth-order　cumulants　joint　estimation　algorithm,　this　paper　proposes　a　new　joint　estimation　algorithm　of　TDOA　and　FDOA　by　using　wavelet　thresholding　denoising　method　combined　with　characteristics　of　non-circular　signals.　The　method　operates　firstly　wavelet　thresholding　denoising　for　the　received　signal,　then　constructs　conjugate　fuzzy　function,　and　finally　two-dimensional　search　is　made　to　obtain　the　time　difference　and　frequency　difference　parameters.　The　simulation　experimental　results　under　different　signal-to-noise　ratio　show　that　the　proposed　algorithm　can　not　only　suppress　correlated　noise,　but　also　has　relatively　lower　computational　complexity　and　also　can　make　accurate　estimation　under　low　signal-to-noise　ratio.　©　2016,　Science　Press.　All　right　reserved.