TL;DR
- Cell-Site Talks: SpaceX reportedly seeks at least 80,000 U.S. cell sites for a wireless network combining ground-based coverage with satellites.
- Indoor Reception: Its proposed 800 MHz network would use radios on towers and rooftops to extend reach and improve reception inside buildings.
- Carrier Access: Research firm MoffettNathanson argues SpaceX would need to pay to use an existing carrier’s network to compete effectively.
SpaceX is in talks with U.S. tower owner Crown Castle for access to cell sites as it develops a wireless network combining ground-based antennas with satellites, Bloomberg reports. Such leases would provide locations for equipment intended to connect mobile users.
People familiar with the private discussions put SpaceX’s nationwide target at a minimum of 80,000 sites. Potential agreements would support chief executive Elon Musk’s stated goal of building a fourth nationwide wireless carrier, adding ground infrastructure to the company’s satellite-based mobile ambitions.
Asked about possible leases with SpaceX and other satellite providers on an October 21, 2026 earnings call, Crown Castle chief executive Chris Hillabrant said: “You can assume that we’re talking to all parties.” He declined to discuss individual customers.
Ground Sites for Mobile Coverage
Tower owners supply physical space where mobile operators install antennas and other equipment. Established U.S. carriers typically rent that space because shared facilities cost less than owning and maintaining their own towers.
In a recent spectrum-transfer application, SpaceX described radios feeding directional antennas on towers, rooftops and other structures. Those radios would provide a ground-based coverage layer for its Starlink Mobile service. The company proposes using the long reach and building penetration of the 800 MHz band to deliver mobile broadband in places higher-frequency networks struggle to serve.
SpaceX and U.S. carrier T-Mobile were collaborating on direct-to-cell connectivity by January 2024. That approach connects ordinary phones through satellites.
Investment bank TD Cowen estimated that a nationwide terrestrial network would cost more than $80 billion, with a buildout taking years.
Customer Locations Shape Small-Cell Coverage
SpaceX has also proposed using its home internet equipment to transmit mobile signals around customers’ installations. These small cells would provide local connections near subscriber premises.
Their location is the immediate problem identified by research firm MoffettNathanson in Bloomberg’s account: customers install equipment where they live, which may differ from the locations needed for consistent mobile coverage. Access to tower sites would give SpaceX another type of location for its radio equipment.
In an August 14, 2026 analysis of the small-cell proposal, MoffettNathanson also pointed to lower antenna heights and power levels, permitting and maintenance. Equipment tied to customers’ rooftops would have to accommodate customers leaving the service and later hardware upgrades.
The analysts also identified a potential advantage in using Starlink for satellite backhaul: it could carry traffic between the local radio cells and the wider network.
More Reach, With Capacity Still Contested
The radio plan rests partly on an October agreement with Grain Management, a digital-infrastructure investment firm, to acquire its nationwide 800 MHz spectrum portfolio. Grain’s announcement described the transaction as subject to FCC approval and customary closing conditions at that time.
The portfolio comprises up to 14 MHz of paired bandwidth in the 800 MHz band. The two figures describe different properties: 800 MHz identifies the frequencies, while 14 MHz describes the amount of radio spectrum available.
This is the last critical piece of the spectrum puzzle needed for SpaceX to provide complete phone coverage in America https://t.co/kKjP072xZf
— Elon Musk (@elonmusk) October 8, 2026
LightShed Partners analyst Walt Piecyk expects 800 MHz to form a terrestrial base layer alongside higher-capacity next-generation satellites. Low frequencies require large antennas, he argued, making their use less attractive on satellites manufactured at scale. SpaceX’s application separately describes seeking authority to add the band to its next-generation satellite system.
Evercore ISI analyst Kutgun Maral sees the relatively modest 14 MHz chiefly as a coverage improvement, rather than a capacity solution. He still viewed the agreement as progress toward a direct consumer service. Reaching a phone and having enough spectrum to carry its traffic are separate requirements.
MoffettNathanson analyst Craig Moffett argued after the Grain agreement that SpaceX would need a wholesale arrangement with an existing carrier to compete effectively. In the earlier MoffettNathanson model, that carrier would provide broad coverage while rooftop small cells handled some traffic locally. The combination could become cost competitive if SpaceX could produce affordable equipment and resolve the operational problems of maintaining it across customer premises.