Interplanetary Network (IPN) Tested

CETS writes "Slim on detail but...USA Today reports of the first test of an Interplanetary Network. 'In a sign of cosmic communications to come, last week mission controllers sent signals to a Mars-orbiting European spacecraft, which relayed the instructions to NASA's Spirit rover on the surface, and a signal was returned to Earth back along the same path.'" NASA also has a press release.

Comments

IPN not like TCP/IP

[Anonymous Coward]

The IPN and the Internet are two different things.

The IPN does not use TCP or another transmission control protocol because it is simply not possible to acknowledge data/rerequest data if the latency is that big (minutes to days in the solar system).

Re:I'm kind of surprised...

[Helvick]

Simple - the primary mission objectives for Orbiters is remote imaging and a low altitude polar orbit is ideal for that because it gives almost total planetary coverage. It means that communications windows with landers are very short (8-12 minutes a day for Odyssey, MGS and Mars Express) but they can cover landers anywhere on the planet at high bandwidth for those communication windows.

This will be the case for the next Orbiter (Mars Reconnaissance Orbiter 2005) and any others prior to the Mars Telecommunications Orbiter which has a primary objetive of being a proper telecoms relay. MTO will provide at least 10x the current bandwidth, communication windows up to 8 hours in duration and will use optical as well as S-Band\X-Band radio links.

Re:Communications Relays

[Detritus]

Phased array antenna systems are used for applications like air defense radar, where you need to track multiple inbound targets, and satellite communications where the satellite transmits to multiple stations on the ground. The difference between a phased array and a parabolic dish is that the parabolic dish is mechanically aimed and can only point at one target at a time, while a phased array is electronically steered and can simultaneously track multiple targets.

A phased array is composed of a large number of simple antennas in a regular pattern. Each of the simple antennas is connected to a phase controlling element, usually controlled by a computer. By adjusting the phase of each simple antenna, the array's radiation pattern can be manipulated to form one or more directional beams, without having to move any mechanical parts.

For NASA's application on TDRS, it allows them to simultaneously track and communicate with multiple satellites in low-Earth orbit, with a single electronically steered antenna system.

The trick NASA pulled with the phased array antenna on TDRS was to take the phase controllers off the spacecraft and put them at the TDRS ground station. The TDRS spacecraft takes the output of all the simple antenna elements and retransmits each one to the ground station. The ground station has a magic phasing/combiner box that takes the outputs of all the simple antennas and adjusts the phase of each signal and combines them under computer control. This splits the phased array into two parts, with part in space (simple antenna array) and part on the ground (phasing/combiner/control computer). This removes a big chunk of hardware and complexity from the spacecraft and relocates it to the ground station.

Looking at the TDRS web page, the latest series of TDRS spacecraft (TDRS-H, I, J) have the beam-forming hardware on board the spacecraft, instead of doing it on the ground.

Reminds me of...

[Beolach]

A comment in the Linux kernel:

/*
* [...] Note that 120 sec is
* defined in the protocol as the maximum possible RTT. I guess
* we'll have to use something other than TCP to talk to the
* University of Mars.
*
* PAWS allows us longer timeouts and large windows, so once
* implemented ftp to mars will work nicely. We will have to fix
* the 120 second clamps though!
*/

--(from /usr/src/linux-2.6.2/net/ipv4/tcp_timer.c, concerning RTT [round trip time])