A Ku-band low incidence backscatter model (KuLMOD) for retrieving wind speeds from Tropical Rainfall Mapping Mission (TRMM) precipitation radar (PR) data is proposed. The data set consisted of TRMM PR observations and collocated National Data Buoy Center (NDBC) and Tropical Ocean Global Atmosphere program buoy-measured wind and wave data. The TRMM PR data properties were analyzed with regard to their dependence on spatial resolution, wind speed, relative wind direction, and significant wave height. The KuLMOD model was developed using incidence angles (0.5 to 6.5 deg) and wind speeds (1.5 to 16.5 m/s) as inputs. The model coefficients were derived by fitting the collocated data. The KuLMOD-derived normalized radar cross section, σ0, was compared with those obtained from the TRMM PR observations and a quasi-specular theoretical model and showed good agreement. With the KuLMOD, the wind speeds were retrieved from the TRMM PR data using the least squares method and validated with the buoy measurements, yielding a root mean square error of 1.45 m/s. The retrieval accuracies for the different incidence angles, wind speeds, and spatial resolutions were obtained.