Can FM Broadcast Antennas Interfere with Cellular Antennas on the Same Tower?

What does Motorola say about vertical antenna isolation?

Motorola's Standards and Guidelines for Communication Sites (R56, 2005), Section 8.4.1, gives a widely cited practical guideline for vertically separated transmitting and receiving antennas:

Motorola specifies the clearance from the tip of the lower antenna to the base of the upper antenna. These figures refer to antenna-to-antenna isolation, not merely the loss calculated from distance in free space. They are useful planning guidelines, not a guaranteed isolation specification for every combination of frequencies and antenna models.

In the example shown here, the cellular panels are positioned approximately six metres above the uppermost FM dipole. This provides a convenient starting point for evaluating coexistence. Read Motorola R56, Section 8.4.1, page 8-9 (PDF).

Why can a two-bay FM dipole array help?

A vertical half-wave FM dipole radiates most strongly outward toward the horizon and has a natural radiation null along its vertical axis. Two identical dipoles stacked approximately half a wavelength apart and fed in phase add an array-factor cancellation mechanism in the upward and downward directions.

At 100 MHz, the wavelength is approximately 3 metres. For a practical two-bay design, the dipole feedpoints may therefore be separated by about 1.5 metres; each physical dipole is roughly 1.4 metres long. The tips of adjacent dipoles consequently sit close together, with a small gap.

In an ideal far-field model, the individual dipole nulls and the in-phase, half-wave-spaced array factor produce very low radiation directly above the array. In real installations, the actual upward coupling is affected by the tower, mounting hardware, feed phases, cabling and the relatively short separation distance.

How do cellular antennas and receiver filters affect the FM signal?

Cellular panel antennas are designed for their allocated cellular frequency bands, whereas FM broadcasting operates between 87.5 and 108 MHz. The cellular panel's response at an FM frequency may be quite different from its advertised cellular-band gain or radiation pattern. Its actual FM-frequency rejection cannot be assumed without antenna data or measurement.

There is a second, distinct line of defence: the cellular receiver's RF front-end filtering. Band-selective filters, duplexers and other receiver components may attenuate the unwanted FM carrier before it reaches a sensitive amplifier or mixer. Motorola R56 specifically discusses receiver filtering to prevent desensitisation from nearby transmitters.

It is important to distinguish RF power coupled into the cellular antenna port from RF power remaining after receiver filtering. Filtering can reduce the latter, but it does not change the electromagnetic signal incident on the antenna.

Example: 500 W FM transmitter and six metres of separation

Consider a 500 W FM transmitter feeding a two-bay dipole array. Each dipole receives approximately 250 W, assuming an equal power split and ignoring feeder losses. We can illustrate the effect of nominal isolation by considering the transmitter as a whole:

This simplified budget is illustrative only. The actual array-to-panel coupling is not established by subtracting a guideline from transmitter power, and 20 dB of receiver filtering is an example, not a claimed specification for cellular equipment.

For this example, a further 17 dB of verified suppression after the nominal 55 dB isolation would reduce the hypothetical +2 dBm antenna-port level to −15 dBm. Such reduction is physically plausible with suitable filtering and other system losses, but −15 dBm must not be described as the typical level at all cellular panels without supporting measurements.

A practical field experience: Ebeye

During a shared-tower evaluation at Ebeye, NTA engineers operated an FM transmitter at 500 W into a single dipole and monitored the cellular data link while switching the FM carrier on and off. No observable change in ping delay or packet loss was reported during that test.

That was encouraging evidence that the tested cellular link continued to operate normally. However, it was not a direct measurement of antenna isolation or LNA input power. Ping tests may miss receiver sensitivity degradation when signals are strong, and the outcome of one manufacturer's equipment cannot automatically be extended to every cellular base station.

More comprehensive commissioning can examine uplink SINR, block error rate, receiver desensitisation, out-of-band blocker power and performance during weak-signal conditions.

What should tower operators take away?

Six metres of vertical clearance is a useful shared-site design starting point, supported by Motorola's long-standing approximate 55 dB antenna-isolation guideline. Two correctly phased FM dipoles spaced at half a wavelength can further reduce upward radiation under suitable conditions, while cellular receiver front-end filtering can provide additional rejection of the FM broadcast signal.

These mechanisms make successful FM and cellular co-location entirely feasible. The precise isolation and receiver input level, however, must be established for the actual installation rather than inferred from a universal guarantee. A properly engineered site combines antenna placement, known receive-path filtering and, where necessary, measured confirmation that the cellular receiver is not being desensitised.