The invention discloses a signal pattern positioning method which aims at a wireless sensor network and does not need a distance measuring technology. For any one target node NTp, firstly, a signal space model is established by using an average value method, and a current network signal space matrix and a physical position information matrix are obtained by calculation; secondly, the receiving signal intensity vector RSSpNT of the target node NTp is calculated according to a signal intensity average value method; thirdly, a neighboring reference position point of the target node NTp is selected by using a compensative K value neighboring method; and fourthly, a three-dimensional coordinate Posp equal to (xp, yp, zp) of the target node NTp is calculated by using a signal intensity weighting method. The method directly uses the signal intensity as a parameter to position, the parameter conversion is not needed, the conversion error is eliminated, the system disposition complexity is reduced, collecting parameters are processed by adopting neighboring selection and weighting algorithm, and the influence of the positioning precision by the node difference, the environmental noise, and the like of the sensor is reduced.