Energy Reconstruction and Spectral Unfolding (Stockholm, June 24,  2002)

Contents

Outline

Reconstruction Methods

ANNE

Comparison

Primary Energy connection

Spectral unfolding

Classification

Limited acceptance

Limited resolution

Instability

Regularized unfolding 1

Regularized unfolding 2

Regularized unfolding 3

Bayesian unfolding

The muon spectrum

The primary spectrum

Outlook and conclusion

E-Mail: geenen@physik.uni-wuppertal.de

Homepage: http://amanda.uni-wuppertal.de/~geenen

Summary: 

I summarize the results shown already at Berkely about neural network energy reconstruction. I present the performances for B10 and AMAII geometries and interprete the reconstructed energy in terms of primary energies. 

I motivate the need of unfolding in spectral reconstruction and describe the method of regularized unfolding. A comparision between Bayesion and regularized unfolding is shown and some preliminary results on spectral unfolding for muons and primaries with AMANDA are given

Conclusion: The performance of neural network energy reconstruction is comparable to standart techniques and spectral unfolding improves with this parameter. Regularized unfolding is more stable than Baysian techiques.  Primary energy reconstruction is in good agreement with expectation in the range from 10TeV -1 PeV, muon energy spectras can be reconstructed in the range from 100 GeV to 100 TeV.

 
Edited with