IT versus WirelessTuesday, September 08, 2009
Historically, wireless networks have been designed and built by people who have skills in both the science and art of wireless. Wireless systems were built with racks of radio equipment, towers, and antennas, and connected back to dispatch or network centers with wired or microwave connections. When there was a problem, a radio technician was dispatched to the site with test equipment, tools, and spare parts.
Today a new battle is brewing. We are using the same towers, antennas, and feed lines, and the radio equipment is still mounted in racks. But the racks are smaller and the connections to the rest of the system are via T1, frame relay,fiber, or microwave links. But when there is a failure, typically, a technician is dispatched to the site after someone monitoring the system remotely has run diagnostics and isolated the point of failure. The technician takes spare plug-in cards with him/her, and "repairs" the system by replacing a bad card or sub-module. Once the part has been replaced, remote diagnostics and a series of tests are run to confirm that the system is back in full operation.
Only a few years ago, we developed site coverage plots by hand, using complex formulas that yielded approximate answers for coverage from specific locations for a specific type of radio service. Today we use complex computer software. We enter the perimeters and the software draws all kinds of fancy plots and we end up with approximate answers for coverage from specific locations.
Today we are heading for a new world where wireless collides with IP and Internet technology. It has been coming for a long time but now it is accelerating and the implications will be interesting. Back when I was helping LA County deploy its microwave backbone for its radio, telephone, and 911 systems, we put large dish antennas on the towers, ran huge cables down to the communications building, attached them to racks full of radio equipment, and interfaced the systems to wired T1 and other types of wired backbone.
Today, someone climbs the tower, clamps a small dish antenna to the side of it, and plugs in an RJ-45 connector attached to a piece of Ethernet cable that runs into the communications shelter where it is plugged into an IP router and then connected to the wired network. By the way, both types of systems work just fine. The newer ones are a lot less expensive to deploy, and when they break they are not fixed in the field, as the older systems were. Boards are replaced until the system starts working again.
IP is changing the world of wireless, and with changes come politics and budget battles. In most wireless organizations today, there is the radio side of the house and then the IT side. Until recently, their staffs did not interact much. The radio engineers, site managers, and technicians are tasked with finding new locations for sites to add capacity or coverage, having the sites built, and then running real-world coverage tests to find out how closely the computer-generated coverage maps match the actual "as built" coverage. They usually find a fair number of differences due to signal shading, signal absorption by trees (pine trees are like sponges for radio signals), and other obstructions either man-made or natural.
The IT department is now charged with getting the output from the site back into the network, although this, too, is a recent change-the communications department used to handle the system end-to-end. This transition has been possible because tasks have become so automated and new back-end systems are being built as all-IP networks. It used to be that radio sites held radios, power supplies, back-up batteries, and not much more. Today, that same site probably has at least one IP router, most of the interconnections are IP-based, and lots of Ethernet cables are running around the site. (These cables have to be shielded and must be grounded to a good ground on both ends to prevent the radios from interfering with the cables, or as is usually the case, to keep the cables from interfering with the radios.)
This is one area in which the crossover between radio and IT professionals is most evident. Most IT trained professionals do not know anything about radio systems, and while most radio professionals probably know a little about local-area networks, they don't know much about how the systems work, nor do they know much about the noise that can be generated by IT systems. Most off-the-shelf network pieces and parts including routers, switches, and the like, are not designed to sit in an environment that has high levels of radio frequency emissions (not dangerous), and most IT boxes are very noisy. This noise can adversely affect the receivers at the site, and any noise generated at the site has the potential to degrade coverage from the site.
As we move into the world of next-generation networks, IP will be the watchword from end-to-end. Yes, radios, cables, and antennas are still required, but most of the other interconnection cables will be Ethernet-based. The good news is that IP is reducing the complexity of the network back-end. The bad news is that special attention has to be paid to noise and potential interference that can cause some very strange problems.
And when there are problems is when the differences between radio-savvy engineers and technicians and IT experts become evident. The average age of seasoned, radio-knowledgeable people is getting older-many who have fought the battles of interference and other radio issues have retired-and few young people are entering this field. The ones who are don't have the training or background to troubleshoot the problems that cause systems to degrade or stop working. It's worth saying again: Interference at a radio site is one of the worst nightmares of communications specialists.
Interference can be generated by local-area networks, by the mixing or two, three, or more transmitted signals (there are complex formulas for some of these mixes), by another radio's internal circuitry, by a controller or a computer, by a loose ground wire, or a rusted fence or bolt. When this type of interference occurs, it takes someone with the knowledge and skill set to work the problem through to identify the cause. Not taking anything away from the IT professionals, but they are not, nor do they want to be, this conversant in the radio world.
The main issue for me is that as the politics of radio versus IP are evolving, the value of the radio experts is not being weighed equally with the value of the IT professionals. The executives who run a company or manage a city or county are not conversant with either technology and they don't understand the issues. Instead, they know that computers and networks are the wave of the future, and that their IT departments will play an increasing role in their systems going forward. This could end up leading to political battles between the two departments or the combination of the two departments with little or no cross training.
Most of the commercial network operators have an executive called the Chief Technology Officer who is responsible for the network(s). Sometimes there is a CTO on the wired side and another on the wireless side, and sometimes the jobs are combined into a single position. This person is responsible for the entire network from the antennas to the back-end systems that make it all work. He or she has to have both radio and IT professionals working under them, and hopefully there is cross training. However, in the cities and counties of the United States there are still two very distinct departments-Radio Communications and IT-and while there is some discussion between the groups, there is also a political battle that could end up with the IT side of the house taking over the functions of the radio communications department.
Every segment of this industry needs to be engaged in cross training radio and IT personnel. It is important as these two worlds come together that there is a realization that neither group can provide all of the expertise needed in today's wireless environment. A battle to see who will come out on top should be avoided. The more I work with various companies and agencies, the more concerned I become about the politics of this situation. This should not be about which department and which "boss" wins, rather it should be about how they can work together to ensure that the marriage of radio and IP goes smoothly and that when a problem develops there is someone with the right skill set available to track it down and fix it.
Andrew M. Seybold