Moon hackers prepare open source communications lab

Alexandru Csete)

The TVRX daughterboard is a complete VHF and UHF receiver system based on a TV tuner module. It can receive a 6 MHz wide block of spectrum from anywhere in the 50-860 MHz range. All tuning can be controlled from software. (Source: Alexandru Csete)

‘Open source’ Google Lunar X Prize contestant Team FREDNET is putting the final touches on the team’s ‘open source communications lab’, which will be used for upcoming prototyping and testing of the radio technologies that will be utilized on the team’s robotic mission to the Moon.

In accordance with the project’s open nature, all schematics and specifications for the hardware used in the lab are freely available on the Internet, allowing anyone with basic electrical engineering knowledge to set up their own lab at home. The software used in the lab, GNU Radio, is available under the open source GPL license and is being developed under the auspices of the Free Software Foundation, which first defined some of the crucial principles that underlie today’s open source community two decades ago.

GNU Radio implements a relatively new technology called software-defined radio. Unlike in traditional ‘hardware radios’, where a number of important signal processing functions are hardwired into the receiver and transmitter with components such as mixers, filters, amplifiers and modulators, in software-defined radio these functions are instead implemented in computer software programs.

“The idea of software-defined radio has been around for many years now, but it is only recently with the availability of cheap high performance AD/DA converters, and fast CPUs that it has become easily accessible,” says Alexandru Csete, team lead of Team FREDNET’s Communications Group, who in the daytime works as a software engineer at Rovsing, an aerospace and defense company based in Denmark. Csete explains that software-defined radios have priceless benefits for space use: first of all, the reduction of hardware means reduced weight, lower manufacturing costs and lower power consumption. Furthermore, the team can do extensive reconfiguration and modification of their communication systems in flight simply by uploading new software to the spacecraft – a great advantage if mission requirements change while the radio is on its way to the Moon and replacing the component physically is just out of the question. Finally, the software-defined radios delivers much better performance for the processing of weak signals – exactly the kind of signals a spacecraft will need to be able to receive when traveling in outer space, hundreds of thousands of kilometers away from transmitters on Earth. “In my opinion, software-defined radio is the only sensible option today,” Csete says when asked if the technology will be employed on the actual mission to the Moon.

And Csete’s team intend to use the open source GNU Radio software not just in the lab, but for the radio platform on board the spacecraft as well. “The GNU Radio library and framework is a really impressive piece of work,” Csete says. “There is a wide range of applications it has already been used for, ranging from simple FM radios to radio astronomy and space applications.” GNU Radio gives the programmer a set of small program chunks called ‘blocks’ which can be combined in a multitude of ways to form a signal processing system fitting the programmer’s exact needs. “While we most certainly won’t use the whole library on-board, we can take the pieces that we need and create our own libraries optimized for the target processor,” Csete says. He also notes that his team is free to extend the framework in areas lacking functionality needed for their mission, thanks to the GPL license which the software is distributed under.

The hardware that Team FREDNET has recently acquired for its communications lab is solely intended for prototyping and testing, however. Although the current setup provides the functionality that will be needed for the actual mission, “the performance is not quite adequate for long range communications and weak signal processing,” Csete says. Individual components of the setup may be reused in the team’s ground stations if they are supplemented with custom-made hardware of the required performance, however, and because the schematics for the hardware are publicly available, the team can in any event use the design as a starting point for the hardware to be built for the actual mission, if deemed worthwhile. Much of the software developed with the lab can most likely be used in the team’s production systems with small optimizations.

an example assembly consisting of the USRP, the RFX2400 and the TVRX mounted in an enclosure.

An example assembly consisting of the USRP, the RFX2400 and the TVRX mounted in an enclosure. (Source: Alexandru Csete)

The lab currently consists of four components, purchased from Ettus Research LLC, a company which also contributes directly to the development of the GNU Radio software, at a price of around 1,800 US dollars. “A lot of money for the private budget, but I expect it will have a good return on investment,” says Csete. The central component is the Universal Software Radio Peripheral (USRP), which connects with a standard host PC through a regular USB 2.0 interface and performs a number of general purpose operations on the incoming and outgoing radio signals, such as conversion between analog and digital representations, allowing the host CPU to operate directly on a radio band. The USRP in turn connects to one of a range of so-called RF frontend components which connects the USRP to the desired radio band. Team FREDNET’s lab currently includes three such frontends – the RFX1200, the RFX2400 and the TVRX – which each covers one segment of the radio spectrum. Both receivers, transmitters and transceivers (a combined receiver and transmitter) can be set up with these components.

“Now we have some building blocks, both hardware and software – think of them as Lego bricks,” Csete says. “Next step is to build a few example configurations and use them in some realistic scenarios. We have a few experiments in mind that we expect to carry out by the end of this year,” he says. The experiments will be announced on the team’s internet forum once finalized.

Csete explains that the lab will support his team in learning about the technologies used in state of the art space communications and evaluating various design options for things such as modulation, forward error coding and signal processing. Furthermore, the lab will be used for testing the team’s own hardware and software implementations for the various communication subsystems to be used on its mission and for defining and documenting a reference ‘lab system’ that can be easily reproduced by other members of the worldwide team or by interested third parties.

“Prototyping, experimentation, and simulations are key elements in a successful space mission,” Csete says. “Some people think it is a waste of time to play around and that we should be focusing on building the ‘real thing’ instead, but believe me, we learn many important lessons during these activities,” he continues and says that this is something that applies to all subsystems, not just communications. The upcoming experiments will help Csete’s team catch up on recent advancements in radio technology and he also intends to create some instructive material and reference documentation on the subject such that some of the experiments can later be repeated as real-time demonstrations during various events.

“I only had some very basic theoretical knowledge of software radios before I started in Team FREDNET – so there was a lot to learn and there is still a lot to explore,” Csete says. “I have used hardware radios since 1990 and I find the software radio world extremely exciting. This is one of the reasons why I’d like our ‘open source communications lab’ to also serve as an educational and experimentation platform,” he continues. “I’d like people with no prior experience or knowledge about the subject to have a chance to learn about space communications. Education is one of our primary missions in this endeavor.”

Since the design documentation for the lab hardware components is freely available in the GNU Radio source repository, interested team members can in principle duplicate the lab at home. “Anyone who has experience with electronics and has access to the required equipment – production of printed circuit boards, soldering equipment, etc. – can build it,” Csete says. “However, it is hardly feasible to build an exact copy of it yourself if you take the cost of raw materials, parts, and required time into account. The advantage of open source in this case is that the schematics is publicly available and you can freely modify it to suit your own needs or optimize it for a specific application.”

Csete has purchased the equipment in the lab for money of his own to get work on the project started: “Until we get sponsorship agreements in place we have to do it like this. It is the same with many other things – servers, network bandwidth, travel expenses,” he says. “For now, I consider the whole project as a hobby and as an investment in the future.” However, he also notes that while the team relies on individual contributions to the extent possible until agreements with sponsors can be arranged, the project has never been intended to be self-financed by team members, nor does the team expect that team members spend their own money on the project.

Should anyone out there be interested in getting involved in Team FREDNET’s communication systems work, Csete advises them instead to get in touch with him directly by e-mail or on Team FREDNET’s forums. “Officially, we are in the initial phase where we capture requirements and do high level analysis,” he says. “One of my top priorities now is to finish a draft system requirements document. This is the beast we will use to validate our final communication system against, so it is very important to get it right and unambiguous if we want to have a chance for space qualification,” Csete explains, referring to the process of ensuring that his team’s spacecraft can stand up to the special requirements of use in a space environment, be allowed on board a launch vehicle, fulfil radio regulation requirements and so on.

“However, I realize the need for ‘activating’ some people who haven’t had a chance to get involved yet and therefore I am about to kick off some software projects that do not require extensive knowledge about space and communications systems,” Csete continues, and states that work packages will be published on the team’s forum as they are finalized. “Finally, those who want to learn about digital signal processing and software radios can follow the instructions on the wiki and install GNU Radio on their PC. They can have a lot of fun without any hardware,” he says.


About this entry