On August 6 the Federal Communications Commission adopted a Notice of Proposed Rulemaking that asks whether devices running under its Part 15 unlicensed rules may communicate with authorised satellites, using the same 2.4 GHz and 5.8 GHz bands that already host billions of Wi-Fi, Bluetooth and proprietary IoT radios. The proceeding covers Earth-to-space transmissions at 2400-2483.5 MHz and 5725-5850 MHz, a combined 208.5 MHz of spectrum, plus possible satellite-to-device links in the 5.8 GHz range. It is the beginning of a rulemaking, not an authorisation for service.
The pitch for manufacturers is obvious. Environmental sensors, agricultural equipment, asset trackers and remote infrastructure monitors could send small amounts of data without carrying a separate cellular or satellite modem. Because the radios already ship in ordinary hardware, the cost argument rests on what is already installed, not new chipsets. That matters in markets where connectivity spend is tight and gateway installation is impractical.
The stakes land here: the FCC is testing whether scarce radio spectrum already serving everyday wireless users can absorb space-based links without breaking what already works. Consumers and industry depend on these bands for Wi-Fi and short-range IoT right now. Opening them to satellite uplinks adds a new use that pricing cannot control, because the spectrum is unlicensed, and nobody yet knows what level of degradation existing users will accept.
It is also not an instant upgrade for the installed base. The FCC document warns that a legal path does not solve the link budget. At 2.4 GHz the commission estimates free-space path loss at about 150 decibels across 300 kilometres, which requires a satellite with a sensitive receiver, a suitable antenna and processing able to pull a weak signal out of terrestrial noise. The endpoint also has to handle Doppler shift, changing propagation delay and short satellite visibility windows, and may need to schedule transmissions around known orbital passes.
Power is the other friction point. Plenty of unlicensed IoT devices run well below Part 15 power limits because their jobs call for short-range links. Pointing them at a satellite would force higher transmit levels, tighter timing and repeated retries, each of which raises energy draw and shortens battery life. Remote devices with no power infrastructure take the hardest hit, and that is the very market satellite IoT targets.
It is worth keeping honest: the likely early use is not broadband through an ordinary Wi-Fi link. It is low-data, delay-tolerant traffic. Periodic sensor readings, equipment status, asset location or emergency messages transmitted when nothing else is available. That limits what the technology is good for in the near term, and it sets the category.
Even within that narrow use case, design changes are still needed. Satellite-capable products may require new firmware, optimised antennas or reconsidered radio designs. Hubble Network precedent exists: the FCC authorised its satellites to receive ground signals matching Bluetooth Low Energy parameters in a narrow 2.4 GHz slice at 100 milliwatts, but under limited operation and a waiver because the band carries no conventional satellite allocation. Moving from case-by-case favour to a broad rule is what this proceeding considers.
Interference is the sharper question. Individual low-power uplinks may be hard to distinguish from ordinary 2.4 GHz traffic, and their cumulative impact on nearby Wi-Fi and IoT networks could rise if many devices start pushing power limits to reach passing satellites. The FCC is asking whether the rule change would encourage more devices to operate near Part 15 limits and whether this would reduce frequency reuse for adjacent networks, authorised services and radio astronomy, including whether geofencing is needed around sensitive sites.
Satellite downlinks carry the bigger risk. A single beam can cover far more ground than a terrestrial access point, and even a relatively weak signal can raise the noise floor across many existing networks or cause listen-before-talk devices to defer transmissions. Options on the table include power flux density limits at the Earth surface, smaller beams, rural or underserved-area restrictions, aggregate limits across constellations, contention-based access, out-of-band emission controls and real-time coordination among operators. Each trades protection against service viability.
Commissioner Anna Gomez supported the proceeding and still warned that the proposals must be grounded in technical studies, given how congested these bands already are. That sums up the tension. Part 15 devices enjoy no interference protection and may not cause harmful interference to authorised services. The FCC is not granting them new protection rights, so the open question is how much terrestrial Wi-Fi or IoT degradation is acceptable before a satellite use stops being a free add-on for existing users.
Unlicensed satellite links, if they reach production, look more like one more connectivity layer than a replacement. Cellular non-terrestrial networks and specialised low-power satellite services still offer managed spectrum, defined service levels and radio designs optimised for long-range work. The unlicensed option trades predictability for lower barriers to device integration, and against a much larger hardware ecosystem it could hold a real position at the low-data end. Read more about connected device coverage.
Hybrid architectures look stronger than any single link. A device can use ordinary Wi-Fi or Bluetooth locally, terrestrial gateways when present and a satellite-compatible mode for occasional out-of-network communication. Another route is a specialised gateway aggregating data from nearby unlicensed sensors and sending it up once. Either model keeps satellite as the exception rather than the norm, and that matches where power budgets actually sit.
Business terms and geography remain unresolved. Satellite operators still have to authenticate devices, manage capacity, route traffic and bill for it, because operating in unlicensed bands changes neither the satellite bill nor the need for device management and security. On geography, the FCC can rule for the United States, but satellite footprints cross borders and no equivalent global framework exists for these bands, so a product designed for US operation may be unable to use the same satellite capability elsewhere.
What the FCC proposed
The FCC adopted the NPRM on August 6, designated FCC-26-51A1, and opened it to public comment through the standard docket. It asks about Earth-to-space transmissions in the 2.4 GHz and 5.8 GHz unlicensed bands, possible satellite-to-device links in the 5.8 GHz range, whether devices could be licensed by rule when communicating with approved satellites, and what safeguards would keep the sharing workable. Those questions sit squarely on the Part 15 framework rather than any operating authorisation, so the outcome remains open pending industry input and technical study. The FCC docket and the FCC-26-51 document are posted at fcc.gov.
Why the link budget matters
The physics leave little room. About 150 dB of free-space path loss over 300 kilometres means a receiver on a satellite has to recover a billion-fold weaker signal from terrestrial noise, and the endpoint has to reckon with Doppler and short visibility windows. For a battery-powered field sensor this means redesigned power budgets, not a software patch, which is why the initial commercial promise is low-data, delay-tolerant applications rather than broadband.
The interference picture
Satellite downlinks worry regulators more than uplinks because a single beam illuminates far more ground than a terrestrial access point and can raise noise floors or trigger listen-before-talk deferrals. Power flux density limits, smaller beams, rural-only coverage, aggregate constellation limits and geofencing are all on the table, and the commissioner who backed the proceeding still flags congestion as a real risk that technical studies must settle. Unlicensed users carry no protection rights, so the acceptable degradation level is essentially the crux of the whole rulemaking.
What comes next
If the record supports it, the commission would move toward rules rather than waivers, and satellite-capable low-data IoT could become a product category rather than an experiment. Manufacturers would still need regional variants because no global framework exists for these bands, and hybrid designs would dominate, since local Wi-Fi, terrestrial gateways and occasional satellite mode fit battery budgets far better than any always-on satellite radio.
The proceeding is a start, not a finish. The next twelve months of filings, technical studies and constellation proposals will show whether it ends with changed rules or with words of caution, and the outcome rides on evidence rather than enthusiasm. Whatever the FCC decides, the 208.5 MHz at stake is already shared by too many devices to put new satellite links on autopilot, and the first real satellite-capable products will depend more on engineering discipline and battery control than on any ruling alone.