In this paper, a working ultrawideband (UWB) microwave imaging (MI) system using basic techniques is proposed to detect and localize a tumor model. The system harnesses the time reversal-decomposition of time reversal operator (DORT) signal processing method. We suggest a modified background Green's function and suitable spiral antennas for time reversal imaging of a breast cancer tumor. In this method, scatterings from a breast phantom are measured. Then, reflected signals are processed using the decomposition of time reversal (TR), known under French acronym DORT, algorithm. In a conventional TR imaging method, free-space Green's function is used for detection. However, in the present method, we employ cylindrical Green's functions of a two-layer medium. This makes a better model of the background medium for TR-DORT imaging. Also, the performance of TR-DORT detector against signal to clutter ratio, and its ability to detect and localize different shapes, sizes, and permittivity values of breast tumor are discussed. Furthermore, a fatty tissue phantom model is used to verify the results experimentally. The MI system is tested to localize a tumor within the breast Debye phantom at the frequency range of 3.1-10.6 GHz. The experimental results show the validity of our proposed system.