Public Safety DAS / BDA Training Series - Chapter 2: Design & Development

Chapter 2: Design & Development

Link budgets, penetration loss, donor selection, and iBwave modeling fundamentals

1
2
3
4
5
6

1. From Planning to Design: The Engineering Handoff

Chapter 1 established whether a building needs ERCES and what standard governs it. Design & Development answers a different question: given the measured exterior signal strength and the building's construction, exactly how many antennas, at what locations, fed by how much amplifier power, will achieve the required coverage? This is an engineering problem with a quantifiable answer — not a matter of experience-based guesswork, even for a designer who has built dozens of similar systems.

The design phase takes three inputs directly from planning: the baseline radio signal survey (Chapter 1, Section 6), the coverage percentage and dBm requirements that apply to this specific building (Chapter 1, Section 4), and the applicable standard's technical requirements (NFPA 1225 Ch. 18 or NFPA 1221, per Chapter 1, Section 2). Everything in this chapter builds on those three inputs.

The Core Design Deliverables

  • A link budget — a calculation proving, on paper, that the proposed equipment chain can deliver the required signal level to every antenna location
  • A propagation model — a floor-by-floor prediction of actual coverage, built in design software, that the AHJ reviews before approving construction
  • An equipment schedule — the specific BDA, antennas, cabling, and splitters/taps selected to achieve the modeled result
Why this phase can't be skipped or shortcut: A system installed without a proper link budget and propagation model is a system built on guesswork — and guesswork fails acceptance testing. Rework after installation (adding antennas, upsizing amplifiers, re-running cable) is dramatically more expensive than getting the design right on paper first. AHJs increasingly require the design package as a submittal before issuing a construction permit specifically to prevent this kind of costly after-the-fact correction.
Knowledge Check: Which three inputs from the Planning & Analysis phase directly feed the design phase?
A) The building's insurance policy, paint color, and elevator brand
B) The baseline signal survey, the applicable coverage requirements, and the governing standard
C) The general contractor's schedule, budget, and change-order history
D) The building's HVAC tonnage and electrical service size

2. Link Budget Fundamentals

A link budget is a running tally of every gain and every loss a signal experiences from the transmitter to the receiver. If the math works out to more signal than the receiver needs, the link has margin. If it works out to less, the link fails — no matter how good the equipment brand is. Every properly engineered DAS design rests on this one calculation, run for both the downlink (donor site → building interior) and uplink (building interior → donor site) directions.

The Core Equation

Simplified to its essentials, a link budget looks like this:

EIRP = PTX − LTX + GTX
PRX = EIRP − Path Loss + GRX − LRX
TermMeaning
PTXTransmit power out of the amplifier (dBm)
LTXLosses between the amplifier and the antenna — cable, connectors, splitters, taps (dB)
GTX / GRXAntenna gain, transmit and receive side (dBi)
EIRPEffective Isotropic Radiated Power — the "effective" power actually leaving the antenna
Path LossEverything the signal loses crossing open air and, indoors, penetrating walls/floors
PRXThe power level that actually arrives at the receiving point

Link Margin — The Number That Actually Matters

The link budget's final output isn't PRX by itself — it's the margin between PRX and the minimum signal level the code requires (-95 dBm, per Chapter 1, Section 4):

Link Margin = PRX − Required Signal Level

A positive margin means the design should pass. A margin of exactly zero is not a safe design — it leaves no room for real-world variables the link budget can't perfectly predict: furniture rearrangement, tenant buildout changes, seasonal foliage growth near a donor antenna, or normal component aging. Experienced designers build in a margin cushion, commonly several dB beyond the bare code minimum, specifically to absorb that real-world uncertainty.

Every loss must be accounted for in dB, not percentage. Splitters, taps, connectors, and long cable runs all cost measurable dB — and because the dB scale is logarithmic, small-looking numbers add up fast. A 3 dB loss isn't "3% of the signal" — it's half the power. Skipping what looks like a minor loss term in the link budget is one of the most common causes of an under-designed system that looks fine on paper but fails acceptance testing.
Knowledge Check: What does a "link margin" of zero indicate about a proposed design?
A) The design has generous headroom and is considered best practice
B) The design is guaranteed to fail acceptance testing
C) The design meets the bare minimum on paper but leaves no cushion for real-world variables
D) The link budget calculation was performed incorrectly

3. Donor Antenna Selection & Placement

The donor antenna is the system's connection to the outside world — typically roof-mounted, it captures the public safety agency's signal on the downlink and radiates the building's amplified return signal on the uplink. Every antenna and every foot of cable inside the building is only as good as what the donor antenna delivers to the BDA in the first place. Get donor placement wrong, and no amount of interior design can fully compensate.

Line of Sight to the Donor Site

Wherever possible, the donor antenna should have clear line of sight to the donor site — the public safety agency's transmitting tower or repeater location identified during the baseline survey (Chapter 1, Section 6). Line of sight isn't always achievable in dense urban environments or where terrain intervenes, but it should always be the starting goal; every obstruction between the donor antenna and the donor site adds path loss that has to be made up somewhere else in the link budget, usually by adding amplifier gain that increases oscillation risk (Section 6 of this chapter).

Directional vs. Omnidirectional Donor Antennas

  • Directional (Yagi, panel): Concentrates gain toward a known donor site direction. Preferred when the donor site location is well established and interference from other directions is a concern.
  • Omnidirectional: Captures signal from all directions equally. Useful when donor site direction is uncertain, or when the system must maintain reception even if the primary donor site changes (some jurisdictions use multiple simulcast transmitter sites).

Avoiding Interference and Multipath

Donor antenna placement also has to actively avoid problems, not just capture signal. Locating the donor antenna too close to rooftop HVAC equipment, other RF transmitting equipment (commercial cellular DAS donor antennas, two-way radio repeaters for building maintenance staff, etc.), or large reflective surfaces can introduce multipath interference or noise that degrades the very signal the system is trying to amplify cleanly. A donor antenna site survey — measuring actual signal quality at candidate mounting locations, not just assuming rooftop-anywhere will work — is standard practice on any serious design.

Donor antenna separation from server (interior) antennas is a design decision, not an installation afterthought. The physical distance and any intervening structure between the rooftop donor antenna and the nearest interior server antennas directly affects the isolation value the system will achieve — which in turn determines how much gain the BDA can safely run without oscillating. This is why donor antenna placement is finalized during design, in coordination with the isolation budget (Section 6), rather than left to whatever spot happens to be convenient on installation day.
Knowledge Check: Why does donor antenna placement matter beyond simply "getting a good signal"?
A) It has no effect on anything besides raw signal strength
B) It affects multipath/interference exposure and the isolation value achievable from server antennas
C) It only matters for the building's Wi-Fi network, not the public safety DAS
D) Donor antennas are always installed inside the building, never on the roof

4. Building Penetration Loss & Material Attenuation

Once the signal reaches the building, it still has to get through walls, floors, and glass to reach interior antenna locations and, ultimately, the spaces where people (and their radios) actually are. Different construction materials attenuate public safety frequencies (typically VHF, UHF, 700 MHz, and 800 MHz) by very different amounts — and the differences are large enough to fundamentally change a design.

MaterialTypical Attenuation at Public Safety Frequencies
Drywall~0.2 dB (essentially transparent)
Plywood~0.3 – 1.1 dB
Standard (non-metalized) glass~0.5 – 2.75 dB
Concrete (normal incidence)Up to ~30 dB
Concrete (oblique incidence, e.g. 70°)Up to ~40 dB — angle matters as much as material
Low-emissivity (low-E) glass~29 – 40 dB
The counter-intuitive result: low-E glass can attenuate worse than concrete. It's tempting to assume glass curtain-wall construction is "easy" for RF and concrete construction is "hard." At public safety frequencies, that assumption is often backwards. Low-E glass contains a microscopically thin metallic coating (applied for energy efficiency) that can attenuate signal as much as or more than a concrete wall. A building that looks like an easy glass-and-steel design on the architectural drawings can be a genuinely difficult RF environment — this is exactly the kind of detail a baseline signal survey and a proper link budget catch, and a design based on assumptions alone misses.

Aggregate Whole-Building Loss

Individual material values are useful for modeling specific wall/floor assemblies, but real buildings are a composite of many materials, air gaps, furniture, and interior partitions. Field studies of whole-building attenuation at public safety frequencies commonly report a mean loss in the high 30s of dB, with substantial building-to-building variation — which is exactly why the baseline exterior survey (measuring the real building, not a generic model) remains the foundation of an accurate link budget rather than a purely theoretical calculation from material tables alone.

Knowledge Check: At public safety radio frequencies, how does low-emissivity (low-E) glass typically compare to concrete in signal attenuation?
A) Low-E glass is always dramatically better (lower attenuation) than concrete
B) Low-E glass can attenuate as much as or more than concrete, due to its embedded metallic coating
C) Glass of any type is completely transparent to all radio frequencies
D) Material type has no measurable effect on signal attenuation

5. The iBwave Modeling Workflow

iBwave Design is the industry-standard software platform for planning indoor wireless systems, including public safety DAS. Rather than relying on hand calculations and guesswork for every antenna location, iBwave lets a designer build a virtual model of the building and simulate coverage before a single antenna is mounted.

The Typical Workflow

  1. Import the floor plan. Architectural drawings (CAD or PDF) are imported and scaled so the software understands real-world distances.
  2. Define material properties. Walls, floors, and glass are tagged with their construction type so the software's propagation engine applies realistic attenuation values (Section 4) automatically.
  3. Input the link budget. BDA output power, cable losses, splitter/tap values, and antenna specifications from the equipment schedule are entered as the RF source.
  4. Place antennas virtually. The designer positions candidate antenna locations on the model.
  5. Run the propagation prediction. The software generates a color-coded heat map showing predicted signal strength across every modeled floor.
  6. Iterate. Where the heat map shows a coverage gap, the designer adjusts antenna count, position, or amplifier output and re-runs the prediction — repeating until the model shows compliant coverage (95%/99% at -95 dBm, per Chapter 1) across the entire floor plate.
The model is a prediction, not a guarantee. A compliant iBwave heat map is what gets a design approved by the AHJ, but it does not replace field acceptance testing (Chapter 4). Real-world results can diverge from the model — furniture, as-built construction differences from the drawings, and equipment installed slightly off-plan all introduce variance. A good design uses the modeling margin discussed in Section 2 specifically to absorb that expected real-world gap between prediction and field measurement.

Deliverables to the AHJ

The output of the iBwave workflow — floor-plan heat maps, the link budget calculation, and the equipment schedule — typically becomes the design submittal package reviewed during AHJ plan review (Chapter 1, Section 5). A design package that clearly shows the propagation model, the underlying link budget math, and the specific equipment proposed gives the AHJ a genuine basis for approval, rather than asking them to take the design's adequacy on faith.

Knowledge Check: What is the purpose of running multiple iterations of an iBwave propagation prediction during design?
A) AHJs require a minimum of five separate software runs regardless of results
B) To identify and correct coverage gaps by adjusting antenna count, position, or amplifier output before installation
C) To calculate the project's total labor cost
D) Iteration is unnecessary — the first model run is always used as-is

6. Isolation Planning — Designing to Prevent Oscillation

A BDA is, functionally, a very powerful amplifier sitting in a feedback loop: it receives signal from the donor antenna, amplifies it, and sends it out through server antennas — but some of that amplified server-side signal inevitably leaks back toward the donor antenna. If too much amplified signal leaks back into the donor path, the system can begin amplifying its own leaked output in a self-reinforcing loop. This condition is called oscillation, and left unchecked, it degrades system performance, can desensitize or damage the amplifier, and can even cause harmful interference to other radio users. Preventing it is a design responsibility, not just an installation-day fix.

Isolation — The Key Design Metric

Isolation is the measured signal-strength difference (in dB) between the donor antenna and the nearest server antenna. The larger that number, the less amplified server-side signal makes its way back to the donor antenna's input. For a stable system, isolation must exceed the BDA's total system gain by a safety margin — commonly cited as 20 dB in current code editions and manufacturer guidance (older references specify 15 dB), though the exact required margin should always be confirmed against the specific code edition and manufacturer in use.

Required Isolation ≥ BDA System Gain + 20 dB (typ. — confirm against applicable code edition)

How Design Choices Improve Isolation

  • Physical separation: Maximizing the distance between the donor antenna and the nearest server antennas — vertical separation (floors below the roof) is often more effective than horizontal separation alone.
  • Intervening structure: A concrete roof deck or additional floors between donor and server antennas adds isolation "for free," since that same building material attenuating outside signal (Section 4) also attenuates the donor-to-server leakage path.
  • Cross-polarization: Orienting donor and server antennas on different polarizations can add several additional dB of isolation beyond distance and structure alone.
  • Directional antenna patterns: Pointing antenna gain away from each other, rather than relying on omnidirectional patterns in both locations.
Isolation is measured and verified, not assumed. A design can calculate expected isolation on paper, but the true value is confirmed with field measurement during commissioning (Chapter 3) — using the same class of test equipment covered in Chapter 4. If measured isolation comes in below what the design assumed, the installed BDA gain typically has to be reduced to compensate, which can reduce achieved coverage below what was modeled. This is precisely why generous isolation margin, planned during design rather than discovered during commissioning, is the cheaper and more reliable path to a stable, code-compliant system.

With the design package complete — link budget, propagation model, donor placement, and isolation plan all documented — the project is ready to move into Chapter 3: Implementation & Commissioning, where this design becomes physical hardware in the building.

Knowledge Check: What does "isolation" measure in the context of BDA/DAS design?
A) The physical distance between the BDA and the building's fire alarm panel
B) The signal-strength difference between the donor antenna and the nearest server antenna
C) The number of floors in the building
D) The battery backup runtime in hours

Final Assessment — Chapter 2: Design & Development

Answer all 10 questions. A score of 80% or higher (8 of 10) is required to pass and receive your certificate.

Question 1: What are the three inputs the design phase draws directly from the planning phase?
A)The general contractor's schedule, the project budget, and the building's insurance policy
B)The baseline signal survey, the applicable coverage requirements, and the governing standard
C)The elevator manufacturer, the HVAC tonnage, and the parking garage capacity
D)The building's paint color scheme and interior design plan
Question 2: In the link budget equation, what does EIRP represent?
A)The exact power level required by the receiving radio
B)The total attenuation caused by building materials
C)The effective power actually radiated from the antenna, after accounting for cable loss and antenna gain
D)The isolation value between donor and server antennas
Question 3: A proposed design shows a link margin of exactly 0 dB against the required -95 dBm threshold. What does this indicate?
A)The design has generous headroom and exceeds code requirements comfortably
B)The design meets the bare code minimum on paper but has no cushion for real-world variables
C)The link budget calculation is mathematically impossible and must contain an error
D)Zero margin is the industry-preferred target for all designs
Question 4: Why is line of sight between the donor antenna and the donor site (the public safety agency's tower) preferred whenever achievable?
A)It is a cosmetic preference with no measurable engineering benefit
B)Every obstruction adds path loss that must be compensated for elsewhere in the link budget, often by adding risky amplifier gain
C)Line of sight is a strict legal requirement in 100% of jurisdictions with no exceptions
D)It has no relationship to the amount of BDA gain the design will require
Question 5: At public safety radio frequencies, which of the following is generally true about signal attenuation through drywall compared to concrete?
A)Drywall attenuates signal far more heavily than concrete
B)Drywall and concrete attenuate signal by roughly the same amount
C)Drywall is essentially transparent to the signal (well under 1 dB), while concrete can attenuate up to ~30 dB or more
D)Neither material has any measurable effect on radio signal
Question 6: Why can low-emissivity (low-E) glass be a worse RF environment than concrete at public safety frequencies?
A)Low-E glass is always physically thicker than concrete walls
B)Low-E glass contains a thin metallic coating that can attenuate signal as much as or more than concrete
C)Low-E glass is a magnetic material that actively blocks radio waves
D)This is a myth — low-E glass never attenuates signal meaningfully
Question 7: In the iBwave design workflow, what is the purpose of the propagation prediction (heat map) step?
A)To generate the project invoice for the client
B)To produce a color-coded, floor-by-floor prediction of signal coverage that guides antenna placement decisions
C)To calculate how many parking spaces the building requires
D)To replace the need for a baseline exterior signal survey entirely
Question 8: A compliant iBwave heat map during design means:
A)Field acceptance testing is no longer required, since the model already proves compliance
B)The design has a reasonable basis for AHJ approval, but field acceptance testing still verifies real-world performance
C)The building automatically receives its Certificate of Occupancy
D)No further design iteration will ever be needed for this project
Question 9: What does "isolation" measure in a BDA/DAS design, and why does it matter?
A)The distance from the building to the nearest fire station
B)The signal-strength difference between the donor antenna and the nearest server antenna — insufficient isolation risks oscillation
C)How many employees are certified to operate the BDA
D)The total square footage of the building
Question 10: Which of the following is a design-phase technique for improving isolation between donor and server antennas?
A)Reducing the number of floors between donor and server antennas as much as possible
B)Maximizing physical separation and using intervening structure, polarization, and directional antenna patterns
C)Isolation cannot be influenced by design decisions — it is fixed by the equipment manufacturer
D)Painting the donor antenna a different color than the server antennas