The telecommunications industry frequently uses a pipe analogy to describe data transport. Twisted-pair copper, used by telephone companies, is considered a narrow pipe; fiber and coaxial cable are called broadband pipes because their bandwidth capabilities are virtually limitless.

While copper, coax and fiber are virtual pipes, the University of Missouri-Rolla (UMR) has proved that real pipes — underground gas lines — can serve as conduits to transmit data using 2.4 GHz off-the-shelf radios similar to the devices used to build Wi-Fi hotspots and home networks. UMR, as part of a U.S. Department of Energy (DOE) grant, sent clear wireless transmissions nearly a mile using a 24-inch gas line and was working on using untethered robots within the line when grant money dried up.

“A big part of dealing with pipelines is the aspect of inspection,” said Richard Baker, project manager with the Gas Technology Management Division of DOE's National Energy Technology Laboratory. “The methodology that they used in this project was a novel approach to communicating to connection devices inside the lines because it actually was looking at using the pipeline as a wave guide for a communications signal.”

UMR originally proposed a secure communications link to remote sensors using the pipeline as a communications medium rather than running a telephone line to it and building a conventional wireline network, said Keith Erickson, department chair and professor of electrical and computer engineering at UMR.

DOE liked that idea but thought it should go a step further. As part of its grant, the agency told the university to expand its approach and use wireless transmissions and mobile robots that would gather the diagnostic data that would be transmitted. So UMR “started looking at commercial off-the-shelf (2.4 GHz radio) stuff,” Erickson said.

“We originally did a study on a section of pipe to see if it was feasible with the frequency characteristics of the pipe. Would it transmit signals in the range that we were looking at?” Erickson said. “It did, and it turned out they were in the range of traditional wireless modems, typically 2.4 GHz.”

The next step was to test the theory on a real piece of pipe.

“We tried it out on a six-inch pipeline we have here on campus, at an experimental mine where we have some property outside the city,” Erickson said. “We made a little pipeline loop, put the normal antennas on them and found that it transmitted pretty well.”

The next step called for a team of university staff and engineering students to try a 24-inch pipeline traveling nine-tenths of a mile. “They got the same results,” proving that the steel in the pipes holds the wireless signals in “like a waveguide … like a big piece of coax [cable],” he said.

This success proved what UMR and DOE officials had believed but no one had ever conclusively demonstrated: A standard 2.4 GHz wireless signal that could feed a residence or hotspot also could be contained and controlled inside an insulated underground pipeline, and it could be aimed bi-directionally to transmit information to receiving points on either end of the pipe. Although some of the wireless data did leak out of the pipe, the majority was held in a very clean, noise-free environment within the pipe itself, thanks to the pipe's insulation and the fact that it was underground.

Once it was determined that a gas pipe could serve as a wireless conduit, the project's second stage called for allowing robots to roam within the gas lines to determine the condition of the pipes and then relay the information wirelessly to receive sites on either end of the line.

“You want the robots to roam down the pipeline, looking at the inside of the pipes, looking for corrosion, potential cracks that haven't gone all the way through but could make a break in the pipe,” Erickson said.

Foster-Miller, a robotic technology firm based in the Boston area, was supposed to develop the robots, but the project never got that far because it hit a financial wall. Although the DOE had been interested in the robots from the start of the project, it just didn't have the money on hand to keep going.

The department, Baker explained, focuses on longer-term, higher-risk research that wouldn't get done or wouldn't get done as quickly if it were left to industry alone. “We don't want any corporate welfare,” he said, adding that the trial “came to its natural end.”

“It was a finite-linked cooperative agreement where they set out tasks that they want to accomplish in a given period of time. They did that, and at the end of that they were free to take this … and put it out in the public for a commercial offering,” Baker said. “Or, if they think it needs more development they can apply again for additional funding, and it goes through the competitive analysis again.”

UMR has sought the additional funding to move ahead with the robots.

“The robots that have been demonstrated have been tethered, always dragging a cable along behind them,” Erickson said. “We're trying to get them to go untethered. That's what the communications link, wireless part, would enable us to do.”

That might happen yet, but it's off the boards right now because, although the project showed promise, it wasn't commercial-ready, according to Baker. “They needed to further test the final range of it, and there were some tweaks that probably could have been done,” he said.

Sending a wireless signal through a gas-filled pipe also should have been tested. But that shouldn't be a problem, Baker said; in fact, the gas would probably improve transmissions.

“As you pressurize the line, you actually can increase signal transmit because it's at a higher pressure,” he said. “We get much better range and sensitivity in a compressed line than you do in just an open air type of thing.”

Regardless of whether robots will work or whether there is a viable commercial application for the project, both Erickson and Baker agreed that UMR has shown the viability of using commercial-grade 2.4 GHz radios to transmit signals using pipelines as conduits.

The project also demonstrated the complexity of the federal bureaucracy and its relationship with academia. Despite the vulnerability of gas pipelines to terrorist attacks, the DOE's study did not take into account homeland security.

“DOT (Department of Transportation) tends to handle security and the safety side of pipelines; our project was focused on integrity, reliability of delivery, that type of thing,” Baker said.

Despite the challenges, the project is far from dead. Baker said that because the initial phase of the project was successful, UMR “should then have the impetus to go forward” with a commercial application Indeed, UMR has had exploratory conversations with an unnamed public utility in Pennsylvania, but things didn't work out, Erickson said.

“We were getting ready to do some testing there, and I couldn't find them,” he said. “I don't know what happened.”

Also, getting the DOT involved could give UMR another revenue source to try to keep the project moving. The university might be able to leverage some work it's already done in the DOT area — and with yet another federal agency, according to Erickson.

“We were involved in some [offshore] natural gas pipeline work and did some reliability studies on some of the sensing systems they had out there,” he said.

Even there, he said, things tended to be complicated from a jurisdiction perspective.

“When [pipes] are under water, they're part of mineral management service with the Department of Interior. You come on land, and they become part of DOT,” he said.

Communication Network

The sensors determining pipeline condition communicate with the central node. They are embedded inside the pipeline but shown externally in this figure for illustration purposes. Each central node communicates with other central nodes using the wireless media of the pipeline as a waveform.