A NEURAL ONLINE TRIGGERING SYSTEM BASED ON PARALLEL-PROCESSING

The study of a prototype of the second-level triggering system for ope ration at LHC conditions is addressed by means of a parallel machine i mplementation. The 16 node transputer based machine uses a fast digita l signal processor acting as a coprocessor for optimizing signal proce ssing applications. A C-language development environment is used for r unning: all applications at ultimate speed. The implementation is base d on information supplied by four detectors and includes two phases of system operation: feature extraction and global decision. Feature ext raction for calorimeters and global decision processing are performed by means of neural networks. Preprocessing and neural network paramete rs rest in memory and the activation function is implemented using a l ook up table. Simulated data for the second-level trigger operation ar e used for performance evaluation.