Proof and Modification of the Burst Generation Rate Model

The burst generation rate (BGR) model is famous for predicting the proton and neutron single event effect (SEE) cross sections of electronic devices based on the heavy ion data, but we cannot find its proof process in the existing literature, and some ambiguities about one parameter exist. In this letter, we give the proof process of the original BGR model, and find that the above parameter should not appear in it. And we propose a modified model by assuming a reaction volume and imposing a constraint on the BGR functions. We carried out the proton single event upset (SEU) experiments for four SRAMs of different feature sizes. It turns out that the modified model performs much better than the original one in predicting the proton SEU cross sections of the four SRAMs.