There are methods and tools for selecting locations for new antennas. There is also the need for a human element, as was shown vividly in a recent Intelsat General Corporation (IGC) case in Hawaii, where, based on a computer search and aerial images, the ideal spot for a 7.3 meter antenna was initially chosen.

To verify a location, a check for potential signal interference that could impede operation of the antenna system is also required.

The computer search of nearby licensed facilities can tell about likely signal interference before one goes to the site. But it remains for a human to inspect the area and to choose where to conduct RF sweeps to determine whether it is suitable for a facility in the proposed frequency bands.

The process involves physically walking the terrain to find physical obstructions. Using a compass and inclinometer or transit level as a guide, one looks at structures or terrain that might impede a signal, to get a ground level view of possible locations for the antenna system.

One such location proved inadequate and was ruled out because of terrain masking. Another site on the opposite side of the facility showed promise - the first step of a two-step test. The second step proved more problematic.

After bringing in a lift to hoist test antennas to the center-line altitude of the proposed antenna system location, a spectrum analyzer was calibrated and connected to the test antenna. The analyzer immediately showed active intermodulation so severe that interfering products continued upward for thousands of MHz.

It wasn't a complete surprise, because an adjacent facility had experienced similar interference for years, but no one understood the source of the problem.

Finding the source of the intermodulation interference first meant determining a pattern to the noise. Tests involved widening the frequency span on the spectrum analyzer until it showed groups of interfering carriers, in this case occurring roughly every 150 megahertz or so.

That meant primary frequencies had to be about 150 MHz apart, and so the investigation continued. Dealing with such a problem becomes a sort of "Sherlock Holmes" process of solving a mystery by eliminating the impossible to find the possible, and ultimately the answer.

More widening of the span showed signals from two cellular telephone base carriers operating at 735 and 880 megahertz. Their peaks were off the top of the scope, coming from sites adjacent to the property and only a few hundred meters away.

Verification came after the cell site operators were asked to do shutdown tests, one operator and one band at a time. None of the shutdowns made the situation significantly better, which meant that the cell sites had clean equipment, and the issue was occurring in the pre-amplifier of the spectrum analyzer.

The signals were so strong that they were saturating a high-gain amplifier, causing the signal to mix in what is called intermodulation distortion. It was clear that this was the major source of interference.

Again, the human element was needed to alleviate the problem. Creating a "notch," or a high-pass filter, could eliminate the strong, unwanted signals that were mixing to cause the interference.

Two notch filters were made using scrap 0.141" hardline parts from a discarded HPA chassis that was at hand. Tuning one of the filters to 880 MHz to meet the demands of the stronger carrier, and the other to 745 MHz to meet the range of both carriers, knocked the strong signals down by 90 decibels, eliminating the intermodulation.

The filters were tuned a bit at a time, winnowing away the issue until their notches were at the correct frequencies, and could push the unwanted signals down enough to cure the overload of the preamp.

It was typical of most troubleshooting efforts which often involve a process of elimination. Interference detection and analysis - both by machine and humans - is no exception to the rule.

Beyond dealing with the cellular base facilities, a terrestrial microwave shot was emitting some spurs on an unauthorized frequency that fell into the downlink band of the proposed antenna system. After the source was tracked down, its owner promised to fix that issue before the IGC antenna was built.

For all of the elements involved in establishing an antenna system, and all of the technology that is involved, it's easy to miss the element that drives the process: the knowledgeable human, walking the ground, using the tools, solving the problems.

It's the key to making sure that the instruments which enable communications can continue to effectively operate. It's also a way in which IGC personnel take the extra step in assuring customers' service quality.