Determining Antenna Separation Requirements for Co-Located Transmitters

Gap

Physical distance between co-located antennas is the primary mechanism for electromagnetic isolation. When two radiating elements share a compact enclosure or circuit board, energy emitted by one antenna transfers directly into the terminal structure of the second. In the reactive near field, defined where the physical distance r satisfies r

Beyond the reactive boundary, within the Fresnel radiating near field, coupling shifts to a spatial atteνation rate governed by $1/r2. Free-space path loss models often fail at close range because the physical dimensions of the antenna elements approach the distance separating them. Predicting isolation accurately requires treating the dual-antenna setup as a two-port network, where the transfer coefficient S21 defines the ratio of received power at antenna two relative to transmitted power from antenna one.

Reactive Near-Field Coupling Mechanics

Near-field coupling in dense hardware integration depends heavily on reactive field structure. Microstrip patch antennas and printed inverted-F antennas (PIFA) store reactive energy primarily in electric and magnetic near fields, respectively. Placing two PIFA elements adjacent to each other on a shared ground plane induces high surface currents across the conducting substrate.

These currents create secondary radiation, bypassing free-space attenuation paths and degrading isolation by up to 15 dB compared to isolated field predictions.

Plane-wave assumptions hold true only beyond the Rayleigh distance (2D2 / λ), where D represents the maximum physical dimension of the antenna. For a 2.4 GHz patch antenna measuring 30 mm across, far-field propagation begins at approximately 14.4 cm. When system layouts restrict physical spacing to 2.0 cm, coupling occurs entirely within the reactive near field.

Orienting elements in spatial quadrature ~ aligning the polarization vector of antenna one perpendicular to antenna two ~ yields 12 to 20 dB of cross-polarization isolation without expanding the physical footprint.

Substrate Wave Suppression and Ground Currents

Dielectric substrates in multilayer printed circuit boards transmit surface waves that bridge physical gaps between trace antennas. FR-4 material, with a relative permittivity (εr) around 4.4, traps electromagnetic energy inside the dielectric layer. This trapped energy propagates as transverse electric and transverse magnetic modes along the ground plane edge, delivering power directly into neighboring RF front ends.

Ground-plane continuity acts as both a shield and a coupling vector. Cutting a slot in the ground plane between two antennas interrupts surface current loops, raising path isolation by 6 to 10 dB. Unplanned ground splits induce slot-line radiation, generating spurious emissions that compromise regulatory compliance.

Engineered defected ground structures (DGS) provide targeted band-stop characteristics, attenuating specific interference frequencies while preserving solid reference planes for adjacent high-speed digital traces.

Quad-polarization layout strategies offer predictable isolation gains when physical enclosure dimensions are fixed.

Harmonics

Reverse intermodulation occurs when high-power RF output from a transmitting antenna enters the radiation aperture of an adjacent transmitter. The injected signal travels backward through the second antenna feed line and impinges directly upon the non-linear active elements of the final power amplifier stage. This non-linear mixing generates intermodulation distortion (IMD) products at mathematical combinations of the two fundamental frequencies, calculated as fIM = |m f1 ± n f2|, where m and n are positive integers defining the order of the product (m + n).

Third-order intermodulation products (2f1 – f2 and 2f2 – f1) fall close to the operating bands of co-located radios. Because power amplifiers operate near saturation to maximize power-added efficiency, third-order intercept points (OIP3) degrade, driving up the amplitude of spurious products. If antenna isolation is insufficient, intermodulation emissions exceed regulatory limits, forcing board redesigns.

A 30 dB spatial isolation between co-located power amplifiers suppresses third-order intermodulation products below the -36 dBm spurious emission limit set by ETSI EN 300 328.

Analytical Reverse Intermodulation Calculation

Predicting third-order reverse intermodulation requires modeling the forward power of transmitter one, the isolation path loss to transmitter two, and the third-order output intercept point (OIP3) of transmitter two’s power amplifier. Consider an industrial gateway hosting a 2400 MHz Bluetooth low-energy radio operating at +10 dBm and a 2420 MHz Wi-Fi transmitter operating at +20 dBm.

Assume an antenna separation distance yielding an isolation S21 of -20 dB. Power from the Wi-Fi module arriving at the output terminal of the Bluetooth power amplifier equals +20 dBm – 20 dB = 0 dBm. The Bluetooth amplifier operates with an OIP3 rating of +25 dBm and a forward output of +10 dBm.

The resulting third-order intermodulation product (2fBLE – fWiFi) appears at 2380 MHz.

Calculating reverse intermodulation power (PIM3) follows the standard non-linear expansion:

PIM3 = 2 PBLE + Preverse – 2 OIP3

PIM3 = 2(+10 dBm) + 0 dBm – 2(+25 dBm) = 20 + 0 – 50 = -30 dBm

The resulting -30 dBm spurious tone at 2380 MHz violates the ETSI EN 300 328 limit of -36 dBm for out-of-band emissions. Fixing this compliance failure mandates increasing antenna isolation to -27 dB, which drops the arriving reverse power to -7 dBm and reduces PIM3 down to -37 dBm.

Mechanisms of Co-Located Spurious Radiation

Intermodulation mechanics generate spurious signals through multiple non-linear junction points across the physical layout.

• Amplifier Output Non-Linearities generate third-order and fifth-order mixing products inside the final stage transistor junction when reverse power arrives from an adjacent transmitting element.
• Passive Intermodulation Dynamics originate at oxidized metal-to-metal contact points, poorly torqued coaxial connectors, and cold solder joints exposed to high local field concentrations.
• Harmonic Superposition Limits emerge when the second or third harmonic of a lower-frequency transmitter lands directly within the active receive band of a co-located higher-frequency receiver.
• Power Supply Cross-Modulation occurs when transient current drawn by a high-power pulsed transmitter modulates the voltage rail of an adjacent low-power RF transceiver.

Passive intermodulation (PIM) creates difficult qualification challenges because it develops outside active radio components. Oxide layers on brass connectors behave as metal-insulator-metal diodes, generating non-linear mixing under field strength exceeding 10 V/m. Eliminating PIM demands torque-controlled mechanical fasteners, non-ferrous plating materials like tri-metal (white bronze), and continuous ground plane soldering along shielding boundaries.

Designing physical layouts without accounting for reverse amplifier coupling leads directly to test-house rejections, extra board revisions, and delayed release schedules.

Blocking

Receiver desensitization occurs when strong out-of-band RF energy from a nearby transmitter enters the low-noise amplifier (LNA) of an adjacent receiver. The high-level signal drives the LNA into compression, reducing small-signal gain and raising the noise floor. When the incoming signal drives the receiver front end into its 1 dB compression point (P1dB), target signals drop below the signal-to-noise ratio required for demodulation, causing packet loss and range degradation.

Reciprocal mixing worsens this desensitization. The local oscillator (LO) inside the receiver carries phase noise sidebands. When a strong unmodulated blocking signal enters the mixer alongside the LO frequency, it mixes with those sidebands, transferring phase noise directly into the intermediate frequency (IF) passband.

This elevated IF noise floor masks low-level signals, degrading sensitivity across the full receiver channel bandwidth.

Can Shielding Cans Replace Spatial Separation on Dense Boards?

Metal shielding cans provide localized board-level isolation, but they do not replace physical antenna separation. Shielding enclosures attenuate radiated energy escaping from board traces, power inductors, and IC packages by 30 to 50 dB, but they do not prevent coupling between external antenna structures mounted on the housing exterior. Radiated emissions leaving an unshielded antenna reach the adjacent antenna directly, bypassing internal shielding.

Shielding cans also alter the cavity resonance of underlying microstrip lines. If a metallic cover sits too close to an unshielded trace, capacitive loading shifts line impedance away from 50 ohms, causing reflection losses and elevating the voltage standing wave ratio (VSWR). Layout rules require placing board shields at a height of at least three times the microstrip trace width to prevent detuning and unintended coupling under the shield perimeter.

Systematic Isolation Audit Steps

Quantifying receiver desensitization requires systematic bench measurement across active operating states.

1. Connect the receiver input port to a calibrated vector signal generator output set to the target technology standard.
2. Establish a stable reference link at 3 dB above the baseline minimum receiver sensitivity threshold.
3. Inject an out-of-band interfering RF tone into the receiver antenna feed port using a directional coupler.
4. Sweep the interfering tone frequency across adjacent operational bands while stepping signal amplitude upward in 0.5 dB increments.
5. Record the interfering power level where bit error rate (BER) or packet error rate (PER) exceeds standard limits.
6. Calculate required spatial antenna isolation by subtracting maximum tolerable blocker power from adjacent transmitter maximum output power.

Compliance with ETSI EN 301 489-17 clause 7.2 dictates that out-of-band transmit power entering a co-located receiver enters below the 1 dB gain compression threshold of the front-end low-noise amplifier.

Internal digital filtering does not remove the need for physical separation; mathematical filtering in baseband processors cannot restore a hardware low-noise amplifier already driven into gain compression by an unattenuated out-of-band RF signal.

Absorption

Co-located transmitters operating within 20 cm of human tissue must satisfy simultaneous transmission field limits. Regulators enforce Specific Absorption Rate (SAR) limits for portable devices operating below 6 GHz and Maximum Permissible Exposure (MPE) power density limits for mobile devices at greater distances. When multiple radios transmit at the same time, their absorption fields overlap, requiring a combined exposure evaluation.

The Federal Communications Commission (FCC) details these routines in KDB 447498 D01 and D04. Devices with concurrent radiators must evaluate the Simultaneous Transmission SAR Test Exclusion Ratio (STER) or perform physical SAR superposition testing. If the sum of individual 1g SAR ratios exceeds 1.0, or if spatial separation between antenna phase centers falls below calculated thresholds, formal laboratory SAR measurements are mandatory.

Simultaneous Transmission SAR Mechanics

Evaluating simultaneous exposure begins by determining standalone 1g SAR values for each co-located radio. For two antennas separated by a distance d (mm), the SAR to Peak Location Separation Ratio (SPLSR) dictates whether physical multi-transmitter testing is required:

SPLSR = frac(SAR11.5 + SAR21.5)d le 0.04

When SPLSR exceeds 0.04, standalone SAR numbers can no longer be summed linearly; full volume scan superposition testing must take place inside an automated SAR chamber. Consider a handheld terminal hosting a 2.4 GHz Wi-Fi antenna (SAR1 = 0.85 W/kg) and a 5 GHz Wi-Fi antenna (SAR2 = 1.15 W/kg).

SPLSR = frac(0.851.5 + 1.151.5)d = frac(0.783 + 1.234)d = frac2.017d

To avoid complex simultaneous SAR chamber testing, SPLSR must remain at or below 0.04:

frac2.017d le 0.04 implies d ge frac2.0170.04 = 50.425 mm

Physical spacing between antenna phase centers must measure at least 50.5 mm to use standalone SAR values. Reducing antenna separation below this 50.5 mm threshold adds 10 to 15 chamber scan days, expanding certification budgets by $12,000 to $20,000.

Antenna phase centers located within the reactive near field merge into a single equivalent radiating structure under regulatory exposure evaluations.

Pre-Certification Multi-Transmitter Criteria

Executing a compliant exposure evaluation requires strict adherence to documented layout rules.

• Phase Center Determination requires locating physical current peaks using near-field scan data rather than relying on board symmetry assumptions.
• Duty Cycle Time-Averaging permits reducing source-based time-averaged output power using standard protocol frame structures under IEEE 802.11 or Bluetooth specifications.
• Enclosure Surface Proximity establishes the minimum distance between human tissue and internal antenna radiators, setting baseline SAR absorption profiles.
• Multi-Band Conducted Power Verification guarantees that module output levels do not exceed upper production tolerances declared on formal FCC grant documents.

According to section 4.3.2 of FCC KDB 447498 D01, simultaneous transmission assessment applies to all co-located antennas separated by less than 20 cm that transmit concurrently, mandating detailed mathematical proof or chamber validation in the formal filing dossier.

Scope

Integrating pre-approved modular transmitters into a single host system does not guarantee end-product regulatory clearance. Certification bodies grant modular approvals based on standalone testing in reference fixtures. When a host incorporates multiple modular radios operating concurrently, interaction effects invalidate standalone grant conditions, requiring formal permissive changes or new authorization filings.

Under Federal Communications Commission (FCC) rules, placing two modular transmitters closer than 20 cm creates a simultaneous transmission environment not covered by standalone modular grants. Integrators must perform a Class II Permissive Change (C2PC) filing under the primary module’s FCC ID or pursue a complete equipment grant under a host-specific ID. Innovation, Science and Economic Development Canada (ISED) enforces equivalent requirements via Class 4 Permissive Change (C4PC) procedures.

Modular Grant Isolation Conditions

Grant condition lines state specific physical operational boundaries. A typical modular grant carries the restriction: “This transmitter is approved for use in mobile and fixed configurations where the antenna can be installed such that 20 cm can be maintained between the antenna and users, and must not transmit simultaneously with any other transmitter except in accordance with FCC multi-transmitter product procedures.”

Integrating a Bluetooth module and a Wi-Fi module into a compact handheld device breaks both conditions: operating distance falls below 20 cm, and simultaneous transmission occurs. Host integrators assume full legal compliance responsibility. The original module manufacturer’s grant protects the host vendor only when integration adheres strictly to declared grant conditions.

Under the European Union Radio Equipment Directive (RED) 2014/53/EU, host integrators follow guidance defined in ETSI EG 203 367. The host manufacturer must execute assessment testing on the final combination to demonstrate compliance with Essential Requirements under Article 3.2 (Effective Use of Spectrum) and Article 3.1b (EMC). Radiated spurious emissions checks are mandatory across active simultaneous modes.

Placing pre-certified radio modules closer than the separation distance stated in their grant invalidates the original modular filing.

Technical Dossier Requirements for Modular Integration

A complete regulatory compliance file for co-located multi-radio hosts requires specific documentation elements.

• Simultaneous Exposure Analysis providing mathematical derivation of SPLSR ratios or complete multi-transmitter MPE summation proofs across all active frequency bands.
• Radiated Spurious Spot-Check Reports detailing physical test measurements taken with all co-located radios transmitting simultaneously at maximum rated output power.
• Antenna Gain Matching Calculations proving that host system embedded antenna peak gain does not exceed the maximum gain permitted on original modular grants.
• Operational Description and Timing Profiles declaring maximum duty cycle constraints and hardware interlocking mechanisms preventing unsupported concurrent modes.

Spectrum regulators demand detailed test evidence when an audit reveals uncertified simultaneous transmission modes inside a custom industrial enclosure.

Margin

Securing reliable RF isolation in dense enclosures requires combining spatial layout methods with passive decoupling techniques. Relying exclusively on physical spacing limits product miniaturization. Hardware engineers implement physical isolation structures on the PCB surface and inside the mechanical enclosure to attenuate coupling modes without expanding device dimensions.

Electromagnetic Bandgap (EBG) structures implemented as mushroom-type periodic metal patches on inner PCB layers synthesize high-impedance surfaces. These surfaces suppress surface wave propagation across specific frequency bands, adding 15 to 25 dB of isolation between antennas separated by less than 0.25 λ. Ceramic notch filters and low-pass LC networks inserted directly into receiver feed lines suppress fundamental harmonic leakage before it reaches active LNA stages.

Field Testing and Chamber Verification

Verification of co-located antenna isolation begins in an anechoic chamber using a calibrated multi-port vector network analyzer (VNA). Measurements capture the full scattering matrix (S-parameters) across all operating bands. Test engineers verify that S21 attenuation satisfies link-budget isolation requirements, adding a minimum 6 dB design margin to account for manufacturing tolerances and dielectric variance across PCB material lots.

Radiated spurious emissions sweeps run while forcing co-located radios into worst-case concurrent transmission modes using automated test software scripts. Rotating the turntable through 360 degrees, alongside antenna mast elevation sweeps from 1 to 4 meters, identifies directional coupling lobes and passive intermodulation hotspots. Catching spurious spikes during pre-certification pre-scans enables minor trace layout modifications before entering formal compliance testing.

Design teams balancing board surface constraints against retest schedule risks select integrated ceramic decoupling structures early in the layout phase, embedding isolation margins directly into the physical layout.