The U. S. Special Operations Forces (SOF) conducted their second Synthetic Theater of War-Architecture (STOW-A) exercise during the week of October 25^th, 1999. The first STOW-A exercise utilized the Distributed Interactive Simulation (DIS) protocol; the second exercise was designed and executed using the High Level Architecture (HLA). The exercise consisted of two missions, which were classic infiltrate and assault using air and ground assets, requiring joint coordination between the Air Force and Army pilots, Rangers and the other SOF units.

This exercise had two primary objectives. The first objective was to conduct a realistic Computer Aided Exercise (CAX) using manned simulators from the 160^th Special Operations Aviation Regiment (SOAR) at Ft. Campbell, KY and the 19^th Special Operations Squadron (SOS) at Hurlburt Field, FL. The second was to establish a distributed training architecture using HLA that could be used to refine and validate tactics for multi-aircraft, all-weather operations. The exercise achieved the goals and objectives of all the participants.

This paper captures the lessons learned during the integration, testing and execution of the SOF STOW-A training exercise using HLA. We address the technical challenges the federation developers overcame related to Run-Time Infrastructure (RTI) connectivity over a Wide Area Network (WAN), and the use of a DIS filter and gateway to integrate radios and non-HLA simulation applications into the federation execution. We discuss the elements used to coordinate and execute the federation between distributed sites.