A novel graphic modeling method of obtaining the fingerprint characteristics of wireless communication channel was provided in the paper. Firstly, discrete Fourier transforms was carried on to the sample data of each channel. And then a three-dimensional image was generated by taking the frequency as X coordinates, the channel number as Y coordinates, and the frequency amplitude as Z coordinates. After the image had been applied to mathematical interpolation, amplitude normalization, extended maximum value filtering, numerical binarization and burr elimination, a binary valued image with fingerprint feature is obtained. Application results showed that this type of fingerprint image can effectively distinguish different scenes of wireless communication.
展开▼
