The JSpOC                                                                                           2 June 2012

The last couple of posts have talked about SSA and how space surveillance data is obtained.  This post will briefly discuss U.S. Strategic Command’s Joint Space Operations Center – the JSpOC.  There are plenty of launch and satellite operations centers across the US and, in fact, around the world.  However, the US has exactly one space operations center that covers the entire gamut of SSA, command and control of US military space forces, and interaction with commercial and foreign space entities.  That is the JSpOC, located at Vandenberg Air Force Base in California. 

The JSpOC has a wide variety of functions, most of which I won’t get into here because they are either dependent on SSA or are aimed at providing command and control of the US’s military space forces.  What I want to discuss is how space surveillance data is taken in and used to develop a space catalog, that is, a complete listing of the position and velocity of any satellite or piece of debris that US space surveillance sensors in the SSN can track. To date, that’s in excess of 22,500 objects.  I’ll start with observations from the radar and telescope sensors arriving at the JSpOC, the analysis that goes into forming orbits, associating those orbits with RSOs already in the space catalog, and finally providing that catalog to a wide variety of users.

Sensors are tasked to track a set of RSOs during a set period – usually 24 hours – so that the RSO positions etc can be updated in the space catalog.  More about the tasking in a bit.  A sensor will take the information it receives at the radar face or the telescope’s focal plane and converts that information into observations – data on the position and velocity of an RSO.  Observations are commonly grouped into tracklets and sent back to the JSpOC for processing, meaning analysis and conversion into orbit data.

A number of things happen to the sensor data – observations – when it reaches the JSpOC.  But first I need to explain a little bit about the equipment at the JSpOC.  The key piece I have in mind is the Space Defense Operations Center computer, commonly known as SPADOC.  SPADOC is the workhorse of the JSpOC in terms of creating and maintaining the space catalog.  SPADOC is an old system – it was originally designed in the 1980s and was finally operationally accepted in 1995 – I know this because I was on the Space Control Center operations floor when it happened.  The system is a set of IBM S390 machines and associated peripherals.  Think “Big Iron” and you’ve got the picture.  Definitely not a modern computing environment – or hardware! – at all.   SPADOC is assisted in its catalog maintenance efforts by on off-line system called CAVENet, which performs some high-precision computations for conjunction assessment and SSN tasking.

So observations and tracklets arrive at SPADOC from the sensors.  The first thing to happen is to ensure that the sensor has tagged the observations to the correct RSO.  Most of the time there’s isn’t a problem but on occasion the association fails and SPADOC corrects it.  Then the observations are used to compute an orbit.  There are as many ways of doing this as there are celestial mechanics (oops – lousy pun!).   SPADOC uses a somewhat brute force method that is quite useful for the quality and amount of data it gets from the SSN – variable and sparse.  In essence, SPADOC uses the observations as input to a series of integration cycles (think back to your calculus – if you never took calculus, consider yourself blessed).  The integration cycles, called differential corrections or DCs, are expected to converge on an orbital solution.  That is to say, the computed error between any given integration cycle and the expected orbit grow smaller and smaller to some arbitrary level, at which point the orbit is considered solved for this round of data.  If for some reason the DCs fail to converge, then a human orbital analyst gets involved and essentially forces the issue by handpicking the observations to be used. 

The upshot of all this is an updated orbit for each RSO.  Keeping the orbits updated is important for military purposes but it also has a very important safety of flight purpose.  Although collisions are incredibly rare – though not impossible as we have seen over the past three years – near misses are more common and every satellite owner and operator wants to know about the, beforehand if possible.  Analysts at the JSpOC run a series of conjunction analyses several times a day – usually once every shift.  Notifications go out to different customers different ways but owner/operators get the updates in the form of Conjunction Summary Messages – CSMs – that are provided free. 

Another product of a current catalog is a new space surveillance network tasking order.  Tasking orders tell each sensor – radar and telescope – what RSOs to look for over a future period – usually the next day – to maintain the orbit.  The tasker takes into account whether a sensor is available, it’s recent performance, and its workload in relation to other sensors in the network.  The programming involved is quite intricate and the process currently takes a few hours but it results in an optimized tasking that doesn’t overload any given sensor, uses the sensors’ unique capabilities in the optimum manner, and as a result gets the needed number of observations to continue to update an RSO’s orbit.

It takes around 200 people to run the JSpOC, performing duties some of which I haven’t mentioned here.  But for our purposes, the JSpOC is the US’s source for SSA.  Military services use it, intelligence agencies use it, NASA and NOAA use it, and a wide variety of commercial satellite flyers use it.  While there are other centers that do parts of the JSpOC job, there is only one JSpOC.  Its products are accepted around the world.  The computer code, which backs this up, is world-class and some parts are considered the “gold standard” for their application.  Billions of dollars worth of satellites depend on its data.  Even it’s competitors use it’s data as the basis for their products – which is fine since no one wants sloppy work leading to a near miss or even a hit on orbit.

This article closes my short series on SSA.  I know that I’ve glossed over the material here, I know that I’ve missed some, and I can only hope that my readers forgive me and give me the chance to fix things.  More to follow . . .