Bright band is perhaps one of the most widely known radar signatures for a long time. Extensive research has been reported in the literature, documenting various features of bright bands. Nevertheless it has always been a challenge to obtain a clear classification of the underlying structure of the vertical profile of reflectivity from radar observations. Classifying the radar profiles of bright band to a limited set will be helpful to further this goal. In this study the Self Organizing Map (SOM) technique is used to classify bright band observed from the Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR). The SOM forms a non-linear mapping of the data to a two-dimensional grid map that can be used as an exploratory analysis tool for generating hypotheses on the shape, and characteristics of the vertical profiles of bright bands. Observation of reflectivity profiles in bright band obtained from the TRMM-PR at 125 m vertical resolution are analyzed and visualized with the SOM. Basic descriptions of the vertical structure are extracted from the TRMM-PR bright band region data, and the corresponding statistics are presented in this paper. Statistics of bright band thickness, peak reflectivity, snow reflectivity, rain reflectivity, and the sharpness of bright bands are studied at a global scale over the tropics. The global classification map provides a unique tool to quantitatively classify bright band region. The methodology and results of the analysis are presented in this paper.