In-Building Public Safety Communications: Coverage, Hardware, and Commissioning
The In-Building Public Safety Communications domain tests ERRCS/BDA work: proving that police and fire radios actually function inside a building, on the AHJ's channels, and that the installation stays inside code while doing it. Expect RF math, code thresholds, uplink/downlink balance, and commissioning documentation.
Coverage and signal thresholds
- Downlink: at least −95 dBm in 95% of the general area on each floor, measured on the AHJ's frequency. Uplink: the donor site must receive at least −95 dBm from a portable transmitting inside the building.
- Delivered Audio Quality (DAQ) 3.0 - speech understandable with slight effort - is the usual pass line; raw strength alone does not prove intelligibility.
- Stairwells, elevator cars and exit passageways are critical areas; many AHJs require the threshold across 100% of that space.
Hardware, licensing and power
- Use an FCC Class B, UL 2524-listed BDA or fiber DAS, installed with the licensee's written consent and registered in the FCC ULS. Consumer Class A boosters are not permitted.
- Donor-to-service antenna isolation must exceed BDA gain by at least 15 dB, or the amplifier self-oscillates and wipes the channel.
- Provide 12 hours of standby battery (the common code minimum), supervised charging, and alarm annunciation where the AHJ requires it.
Commissioning and re-testing
Test on a uniform grid (commonly at least 20 locations per floor) plus every critical area, using the AHJ's own channel and a calibrated, DAQ-capable meter; record date, frequency, location and reading.
Worked example
Donor antenna receives −90 dBm. Jumper and coax loss 3 dB gives −93 dBm at the BDA input. Downlink gain of 80 dB brings it to −13 dBm at the output. Splitter and distribution coax lose 20 dB, antenna gain is 2 dBi, so EIRP is −31 dBm. Worst-case path loss of 55 dB delivers −86 dBm at the portable. Margin is 95 − 86 = 9 dB, so the design passes with room for body loss.
Traps
Mixing dB and dBm; omitting jumper loss; confusing dBd with dBi (2.15 dB); designing downlink-only and later finding uplink noise; testing with a scanner instead of a DAQ-capable meter; and claiming coverage without listing grid test locations.