Method A involves training two Neural Networks, one for all RISE operations, second for all FALL operations with LearnVectors obtained in Method. This Neural Networks are beeing later used during processing live data to determine if to proceed with RISE or FALL operation.