Pulse Shape Discrimination of n-γ Based on Pulse Shape Model and Adaptive Kalman Filter

Motivated by drawbacks resulting from current neutron(n)-gamma(γ) discrimination methods, this paper proposes a new method of n-γ discrimination on the basis of pulse shape model of scintillation detector and principals of adaptive Kalman filter. The proposed method extensively exploits model parameters' differences between n and γ ray. By properly establishing the state and measurement equations of linear Kalman filter based on pulse shape model with three exponential terms, the track framework of pulse signal is developed for a novel indicator of model residual denoting either the pure noisy component or a combination of model bias and noise. 7056 pulses signals sampled from a 252Cf source are studied for the performance comparison of discrimination between proposed method and other state-of-art methods. The qualitative and quantitative results demonstrated that the proposed method has better capacity of n-γ discrimination.