This thesis investigates the performance of the so called muon filter, a critical component of the MUonE experiment aimed at precisely determining the hadronic leadingorder (HLO) corrections to the muon’s anomalous magnetic moment. The filter, located downstream of the electromagnetic calorimeter, is designed to identify muons and reduce hadronic contamination, particularly from pions. Using simulations conducted with FairMUonE software on the Bura supercomputer, various absorber layer configurations were analyzed to optimize the filter’s performance. Key findings reveal that increasing absorber thickness does not impact the acceptance of the muon detector significant enough to compromise detection efficiency. The analysis also demonstrated that the distinct angular distributions of reconstructed muons and background particles enable the development of the Angle Method, which effectively identifies muon tracks by applying a cut-off on angular distributions. This method shows a success rate exceeding ∼ 99% in distinguishing muons from background noise. Furthermore, the simulations confirmed the essential role of the muon filter in enhancing particle identification, especially in ambiguous regions where muon and electron angles are similar. Without the muon filter, up to ∼ 80% of events in these regions could not be identified. The results underscore the necessity of the muon filter for accurate measurements in the upcoming 2025 test run, providing valuable insights for future experimental setups and strategies.