Multi-turn Hits Separation In MDC Of BESⅢ With Neural Network

The Beijing Spectrometer Ⅲ(BESⅢ)has been running at the Beijing Electron Positron Collider II(BEPCII)since 2008,which is designed for the precise measurement and the search of rare decay and new physics in τ-charm energy,and haved achieved remarkable physical results.With the improvement of the luminosity of BEPCII and the data statistics,higher quality of data processing and better offline software are needed.The tracking detector of the BESⅢ is a Main Drift Chamber(MDC).The tracking efficiency and the quality for the drift chamber of the BESⅢ experiment is essential to the physics analysis.The reconstruction software of drift chamber is the core part of offline software system.High reconstruction efficiency of charged particles with a wide range of momentum is our physical target for the MDC reconstruction software.Now,the tracking efficiency for the high transverse momentum is high but still has room to improve for the low transverse momentum tracks,especially for the tracks with multiple turn.This is because multi-turn tracks,which with the hits from different turns will adjacent to each other and even overlap will be leaved in MDC due to the magnetic field and the structure of detector.And it hard to separate the first turn’s hit from the non-first’s turn hits for the multi-turn tracks.A novel way to use a convolutional network called U-Net network is represented to solve the identification of the first turn’s hits for the multiple-turn tracks.The method can well realize the separation of multiple-turn hits,which has been successfully applied to the BOSS software system.