Abstract
This work presents an extensive study of signal detection in the presence of noise for a high-energy calorimeter used in particle collider experiments. The objective is to select the calorimeter cells, out of approximately 10,000 available readout channels—most of which contain no signal—that should be considered for energy reconstruction. To achieve this, several signal detection techniques are explored, including Maximum Likelihood, independent component analysis, and neural processing. The results indicate that the neural network approach outperforms the other methods, both in terms of detection performance and implementation complexity.