Comments on Optimal Multiple Level Decision Fusion with Distributed Sensors