Abstract: K-M(Keenan-Motley)model adds the loss of the single wall to calculate the total loss of indoor wireless signal penetrating multiple walls,which exists errors .To solve this problem,a variety of indoor scenes are selected for CW (Continuous Wave)tests,and the influence factors and changing rules of wireless signal through the wall loss are analyzed.An indoor wireless model through the wall loss correction method based on artificial neural network is proposed.The test data are trained and then the prediction model is established.The experimental results show that the predicted model conforms to the model correction criterion and has good prediction accuracy in the actual indoor scenarios.

Key words: K-M(Keenan-Motley)model, CW(Continuous Wave)tests, Artificial neural network, Model calibration criteria