Cornell scientists have cracked the codes needed to track navigation signals from a prototype Galileo positioning system satellite, leading to a scuffle with the EU-based project, which intends to keep tight control over access.

Earlier this year, a team led by Mark Psiaki, professor of mechanical and aerospace engineering at Cornell and co-leader of Cornell's GPS Laboratory, devised a way of determining the pseudorandom noise (PRN) codes needed to track its navigation signal, which hadn't been released at the time.

The team published

their findings in the June issue of GPS World.

To Psiaki's consternation, the crack doesn't appear to sit well with the strict licensing terms the Galileo project is putting on the codes, in an attempt to keep control over which devices are permitted to track Galileo signals. That's in spite of the fact that the codes are "open source", according to Psiaki.

"Apparently they were trying to make money on the open source code," he said in a statement.

Galileo is backed by the EU as an independent alternative to the US' Global Positioning System (GPS), which is primarily intended for the military, and as recently as 2000 only provided limited access to the public.

The 3-billion euro Galileo project is expected to be in operation by 2010, with the aim of providing better accuracy, improved coverage at higher latitudes and independence from the US system.

While the EU and the European Space Agency are Galileo's prime movers, the project is three-quarters privately funded, and needs to make a profit.

The project is planning measures such as an encrypted version of the service (called Commercial Service or CS) that will deliver improved accuracy - better than one metre, or, with the addition of ground stations, down to less than 10 cm.

However, Galileo won't be strictly fee-based - an Open Service (OS), freely available to any Galileo receiver, will also be provided, with accuracy within four metres. Manufacturers will, however, need to licence PRNs from Galileo in order to manufacture OS-capable receivers.

Psiaki targeted GIOVE-A (Galileo In-Orbit Validation Element-A), the prototype for the 30-satellite Galileo system, which went into orbit late last year. As of January, the PRN codes hadn't yet been released, except to a handful of authorised testers.

"Then it dawned on me: Maybe we can pull these things off the air, just with an antenna and lots of signal processing," Psiaki stated.

His team took about a week to develop a basic algorithm to extract the codes, using codeless acquisition and statistical signal processing techniques. It then took them until mid-March to fine-tune the details.

The team published the codes on their Web site on 1 April, and a few days later, a Canadian manufacturer of GPS receivers used the codes to successfully track GIOVE-A signals.

Galileo finally published the codes in mid-May, as part of its Interface Control Document (ICD), but included intellectual property restrictions that struck industry observers as surprisingly strict.

"Use of the ICD is permitted for research and development purposes for the benefit and the promotion of European GNSS programs. Only time will tell how many manufacturers will elect to engage with Galileo under these rather restrictive terms," wrote Alison Brown, president of GPS receiver maker NAVSYS, at the time.

The PRN codes aren't high-security information, but the experience so far doesn't bode well for Galileo's approach to releasing information - or its success in keeping sensitive information secret, according to some observers. "Security by obscurity: it doesn't work, and it's a royal pain to recover when it fails," said security expert Bruce Schneier in a blog post last week.