Many years ago we were trying to plot the polar diagram of a 90 foot satellite dish by firing a tiny signal from a convenient tower some 20 miles away and panning up/down and left/right. However, we could never get stable signals - always a slow up and down drift of the signal level by a fraction of a dB over several minutes. Then we realised we were looking across Penzance Bay and it was the tide rising and falling - classic interference, even though the water level was theoretically well below the Fresnel zone. Curiously we also realised that with such a huge dish, even 20 miles was technically not far from "near field" at 4GHz. The best way of checking the dish G/T was to do a spot of radio astronomy!
Wonder about the origin of the 1hz vs 2hz AM rate difference between the passing passenger train vs. the freight train? I note that a standard passenger rail car is 85 feet long, while a standard freight car is 44 feet long. Perhaps there is some connection there, as the breaks in the amplitude modulation may be tied to the differing time constant of the gaps between trains with longer cars vs. the shorter cars. Perhaps they were making and breaking the multipath at each space between the cars?
A higher-gain antenna would be helpful; it would not only increase the signal strength of the desired, direct signal but would also help to discriminate against the multipath signal(s). If a pattern null could be directed to the direction of the train so much the better. A small change in height of the antenna might also be useful. Care would need to be taken in using a shield lest diffraction effects actually increase the undesired signal.
This sounds like a classic case of Fresnel Zone interference. When I was researching microwave paths for TV station studio to transmitter and remote pickup links, I always had to determine what would be in the Fresnel zone or transmission path problems could cause problems down the road. Usually the obstructions in the Fresnel zone were stationary, but sometimes not, as in this case.
I once worked at a TV station where the studio to transmitter microwave link path passed across the Mississippi river. The path was only about five miles long, so signals were usually quite strong. However, occasionally we would get a partial fade out of the signal. Something we would have a complete loss of signal for a few seconds. Naturally this would happen during a dramatic point in the action of a story or worse, during a daytime soap opera! We received many irate calls from our viewers. This went on intermittently for a few years, and I was tasked to investigate the problem.
I determined that the problem only started after a second bridge was built over the river about five years earlier. Further, the problem was the worst during the spring at times of high water in the river.
I eventually figured out that the center of the beam of our microwave link path was only about 80 feet above the high water level in the river. The microwave path passed under both of the bridges, just to the right of a support column for the original bridge, and then just to the left of a support column for the second bridge. We were shooting through a keyhole about 200 feet wide and about 80 feet high that was placed directly in the middle of our signal path!
Whenever the water was high and a large ship crossed the path, it would partially obstruct the signal for a few minutes, sometimes completely obstructing the signal and causing a complete drop out. Because the bridge support columns effectively blocked most of the Fresnel zone signal, there was nothing to prevent a complete loss of signal when the main beam was obscured.
To document what was happening for management, I connected a chart recorder to the AGC line of the microwave receiver. I stationed an observer on the tower with binoculars and a radio to alert me when a large ship passed downstream. As ship passed through the link path, I was amazed to see the pen of the chart recorder clearly delineate the bow of the ship, the wheelhouse, the stern, and the trailing wake of the ship on the chart paper.
We eventually rerouted the microwave path via a dogleg path using a taller building in the downtown area. After that, the signal remained stable and the calls from our viewers ceased.
Any type of moving traffic can and does cause similar effects. "airplane multipath" or "airplane flutter" is a well-known phenomenon which affects FM radio and both analog and digital TV reception. I am certain that many of us have noticed our auto FM radios fading in and out when another vehicle moves nearby when you are stopped at a traffic light.
This anomaly is just remarkable to me.Understand the dynamics of traveling signals, and consider this: The mere fact that literally 1,000's of commuters are simultaneously accessing any one particular Cell Tower at the same time; and then consider that each of the individual subscriber signals are dynamically "handed-off" to next cell tower, (usually about 7-10 miles away) as they zip down the Interstate at 80mph; and last, to consider that each of these dynamically changing subscriber signals are not interfering with each other, nor any of the other of the 1,000's of cars from which the signals originate (being an avg. of 4,000 pounds of steel moving at 80mph) is a marvel of 21st Century technology that just about everyone takes purely for "granted".With that degree of RF precision executing routinely in everyday life, I'm struggling to understand exactly how the train on the horizon affected the signal integrity; but the author's rooftop observations seem to correlate the evidence.Hard to believe...
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.