In positron emission tomography a scatter estimate is calculated prior to image reconstruction and is then added at each MLEM iteration to the estimated true counts to yield the estimated prompts. The absolute scale of the scatter is usually estimated by matching the scatter estimate to the data for those lines of response that do not intersect the patient. The goal of this work is to explore alternatives to this tail fitting method. Pursuing the method introduced by V. Panin at the 2012 IEEE Medical Imaging Conference for the case with known attenuation, we estimate the scatter scaling factor as an additional voxel, which is updated at each iteration at the same time as the activity. This method is extended for applications with unknown attenuation. This preliminary study uses a single-scatter estimate and simulated 2D data.