Detailed Description of DSU Experiment

The DSU experiments involved connecting the Kellie Chouest (a US Navy ship) to a land-based Internet connection. The boat was scheduled to be relatively close to the pier (within 5 miles). This eliminated line of sight issues and allowed us to overpower any possible reflections from the waves.

The setup involved connecting a 14.5 dBi yagi antenna to a lucent access point on the Scripps Institution of Oceanography (SIO) pier. The access point was connected to the SIO network which is connected to the Internet through UCSD and SDSC's connectivity providers.

The local router at SIO was talking to us with multicast but unfortunately it did not have proper upstream multicast connectivity. As a result a portion of the experiments were limited to the local area. However, users were able to connect to the onboard camera (Ricoh i-700, 3.3 Megapixel, web controllable) from around the country.

We had a station on the deck of the ship with yagi antennas on tripods. This is where we broadcast the undersea video and connected the onboard camera. The station designer/operator was Hans-Werner Braun. The yagi antennas were used without amplifiers to increase the signal to noise ratio which would reduce the retransmissions and allow for better multicast. However, we found that stand-alone PC cards in the side of laptops seemed to work pretty well.

The underwater video feed was received from the control room in NTSC format and we used a laptop with a USB frame grabber to multicast the video. Due to issues with SIO's multicast configuration we were unable to make these sessions travel across the mbone. However, after the multicast was rebroadcast from the pier it was easily viewable in the navigation room of the ship.

The onboard camera was a 3.3 Mega-Pixel Ricoh i-700. It has an Ethernet card but no wireless card, so we used a lucent WaveLAN/EC to make it talk on the wireless network. This camera had a web server where people could see current images of the on-deck activity and shore (as well as capture 3.3 megapixel images for later use).

These two devices allowed us to transmit video to shore. Additionally, we had another setup in the navigation room of the ship where we could watch the multicast video and the onboard camera. This setup consisted of simple laptops with 802.11b cards. However, in the later parts of the experiment we journeyed further from the pier and we were no longer able to connect from inside the Navigation room. However, we were still able to watch the video from standing outside the navigation room. Had we used a yagi antenna we would have easily been able to continue watching the video from the navigation room.

The diagram shows a navy crew member watching the video and interacting with the onboard camera from the navigation room. The remaining photos are available.

During the dive the boat's position was controlled by GPS. In other words the GPS system controlled the drive propellers and an additional rotating propeller to keep the ship in its proper orientation and position. This allowed us to setup directional antennas on the boat without worrying about changing orientation and position. However, due to the size of the boat its change in position would have been slow and we could probably have tracked it with a gyro and an antenna rotor. However, that is the subject of future experiments.

This document was created by: Todd Hansen (ROADNet Systems Coordinator) with the assistance of the ROADNet team.