The economics and technologies of sharing cellular tower space are
about to undergo major changes. Is your company prepared?

Hidden inside this fake tree is a cell tower from Glen Martin.

Until the 1990s, an operator could install a wireless tower almost wherever it pleased and with minimal objection from local communities. It didn’t take many towers to support an entire region because there weren’t many customers in the first place and most of the data was simple voice. Then towers grew taller, bringing concerns about everything from aesthetics to radiation. Now, with data services proliferating and citizens more astute than ever, there are new trends emerging in tower equipment co-location.

Electronic interference and tower design are among the major issues, both being consequences of other technology evolutions. There’s also cost, an issue that never goes away.

Despite a regular stream of government agencies worldwide reassigning spectrum bands, problems with interference per se are not getting any better or worse than a few decades ago because physics doesn’t change, noted Kevin Linehan, chief technology officer for base station antenna systems at CommScope’s Andrew Wireless Solutions division.

“There’s nothing really new in the antenna technology that’s changing the coupling between the antennas. It’s still based on the coupling you expect between a linear array of dipoles,” he said.

What does change is the horizontal and vertical separation between equipment, which in turn requires tuning. “It’s generally not something you can measure directly in the field unless you have a lot of equipment,” Linehan explained. “For horizontal spacing, the electrical isolation is greater than 35 dB, as little as 12 inches. For vertical separation, the isolation is up to 50 dB with as little as 6 inches of separation.”

MIMO & RET
The numbers change when multiple-in, multiple-out (MIMO) technology enters the equation. Andrew conducted research with the University of Texas at Austin and found that dipole antennas (two arrays inside one radome) using 2x2 MIMO need separation of 3 lambda or more to get correlation in the channels. That means less cost for operators and less resistance being required against wind load. That, in turn, makes life easier for co-location purposes. Diplexing the physical cables that enter a tower also can help, and Andrew is currently developing that technology for modern bands such as 700 MHz and 2.6 GHz, he said.

Linehan: Interference is a major concern.

Something else that can simplify co-location is remote electrical tilt (RET). A technician can disturb his own company’s antenna or a competitor’s antenna any time he climbs a tower. Although networked remote control technology lets many routine tasks such as safety checks and position adjustments happen from a ground service center far away, some tasks must be done in person.

In addition to safety, the cost savings are clear – no need for a large fleet of trucks and lesser needs for insurance – all being factors that make tower-top neighbors happy. The technology can be used for ordinary tuning and for disaster recovery, he said.

Self-optimizing networks is something Linehan said Andrew is working toward. Some forms of RET are in the market today, but mainstream acceptance is 2 to 4 years out, he said.

DESIGN
Co-location is also a factor in designing the physical towers. Chris Martin, director of engineering at GlenMartin Engineering, said his company’s short-term emphasis is to design towers that satisfy municipal rules. Often, that means “monopine” systems, which look similar to real pine trees. Demand is increasing because zoning boards are not allowing new traditional towers. In some cases, carriers actually have had to replace old towers with new visually pleasing designs, he said.

Martin: Construction matters to co-locators.

Monopine towers do not use steel, so special considerations exist for strength. “We understand where they’re coming from and are glad to supply those systems to our clients. Our endeavor has been to streamline the materials that we utilize. There’s a lot of education that has to go on in the industry in the next year or two,” Martin said. He’s referring to carrier infrastructure budgets which in many cases are developed using cost estimates from obsolete simple metal towers.

Martin also said geography is a huge concern. Pine-style towers work well in Florida, but what about Minnesota? “The difficulty in all this has been that we have to serve a very, very diverse geographical area here in the United States. It’s not so much that the clients themselves want anything different; it’s actual geographic location that makes things difficult,” he said. It’s not just appearance, but also subjects like ground faults, rain levels and wind loads that differ across a country as large as the United States, he added. Carriers are just beginning to adhere to Telecommunications Industry Association standards developed in the past few years, he said.

As with the interference issues, tower construction affects co-location because each carrier must deal with limitations to how much equipment can fit in a single space. To help, GlenMartin is researching so-called “smart” systems, using new materials other than steel and that can actually adjust themselves based on sensing the real-time environmental conditions, Martin said.

Sobieck: You scratch my back, I’ll scratch yours.

Nsight, a small carrier in the Wisconsin area, participates in renting space from other carriers and from tower companies, along with renting out its own spare tower space. “One trend is ‘more.’ There’s more of everything. They want more antennas on the tower, they want more cable to hook to the antenna,” along with more footprint and more backup power, said Bob Sobieck, director of engineering. “The last couple years I’ve been seeing this to be a trend.”

With upcoming technologies such as Long-Term Evolution, “I see people already hedging to make lease arrangements, to get assurance that they can operate these services in the future. And we’re guilty of doing it too, because if you don’t plan it now, you’re in trouble later,” he said.

Sobieck likens co-location to a cautious dance because you must play well with your competitors. How you treat a company co-locating on your tower could affect your future. You may need to use their towers tomorrow, Sobieck noted. With typical co-location lease prices of $1,200 to $3,000 per month, and with so many traditional towers soon to need upgrades, technicians do not have a choice. “We’re trying to balance cost vs. aesthetics. You can really throw a lot of money at that stuff.”

DAS IS NIMBLE

When traditional cellular towers can’t be used because of site or zoning problems, an alternate option is distributed antenna systems (DAS). The idea behind DAS is to install antennas on common structures such as telephone poles and street lamps, powered by remote base stations and connected with fiber optics. ExteNet Systems, NextG Networks and NewPath Networks are some of the big names in this field. “Initially, it was a solution in areas where the carriers had a hard time being able to build towers. Instead of building a 200-foot tower, you utilize existing locations,” said Tormod Larsen, vice president and chief technology officer at ExteNet. The tradeoff is cost. Carriers must pay to install the fiber and more nodes are needed to cover the same territory as a traditional tower. Also, some zoning boards or land owners demand custom installations to blend in with existing scenery. However, sometimes carriers can save money by integrating a DAS system for outdoor and indoor coverage with just one set of base stations, Larsen noted. “It’s more a mentality shift than really a technical tradeoff,” he said. However, co-location with the smaller nodes does become a technical challenge, because smaller structures can’t hold as much equipment as full towers. So companies such as ExteNet are constantly encouraging network equipment makers to design products that are small and reconfigurable, Larsen said. Even when several competitors’ antennas do fit in a single DAS node, there can still be issues related to electronic interference and spacing. Looking ahead for DAS, “it’s not replacing the towers, but it’s definitely a complement and you will see it become a tool in capacity planning from the get-go rather than being an afterthought,” Larsen said. “With that infrastructure in place, we could do other stuff as well that supports wireless networks,” such as backhaul services. “That’s definitely a path that we are going to and it’s a very natural thing for us to do.”