Abstract: The automatic deformation monitoring system of TM30 measuring robot based on VB language is developed，and the composition and data serial communication mode and data acquisition mode of the system are analyzed in detail.In order to improve the accuracy of the observation data，the observation data processing methods are discussed.Firstly，the original observation data is conducted on differential treatment so as to weaken the effect of public errors，Secondly，the deformation data signals is conducted on 3 layers wavelet decomposition based on the wavelet analysis toolkit of MATLAB software，and the wavelet decomposition coefficients are processed by soft threshold to weaken the effect of noise.Performance of the deformation monitoring system is tested by adopting the actual deformation monitoring data of Huaibei mining area，the test results show that，the accuracy and stability of the monitoring data are improved based on the processing methods of differential treatment and wavelet denoising，so the method can provide some reference for dealing with the deformation monitoring data of mining area.

Key words: TM30 measuring robot, Deformation monitoring, Differential treatment, Wavelet denoising