Networks & InfrastructureStarlink MobileQualcomm5g InfrastructureDirect To Device Satellite

SpaceX signs Qualcomm for tens of thousands of Starlink Mobile sites

SpaceX will deploy tens of thousands of Qualcomm-powered 5G sites for Starlink Mobile's planned 2027 nationwide US launch, in a deal sized to rival major operators' networks.

4 min read

SpaceX taps Qualcomm to power Starlink Mobile in USNetworks & Infrastructure
SpaceX taps Qualcomm to power Starlink Mobile in USmmisof / pixabay

Why it matters

  • SpaceX signed a multi-year deal with Qualcomm covering tens of thousands of Starlink Mobile cellular sites in the US.
  • Starlink Mobile targets nationwide deployment in 2027.
  • Qualcomm's Durga Malladi said planned installation counts will rival those of major operators.
  • The deployment uses Qualcomm Dragonwing Massive MIMO radios, baseband and software for terrestrial 5G.
  • The companies also agreed to collaborate on AI-native 6G infrastructure, with no timeline disclosed.

The story

SpaceX has committed to deploying tens of thousands of cellular sites under a multi-year deal with Qualcomm, the clearest signal yet that its planned 2027 nationwide launch of Starlink Mobile in the US is a serious infrastructure build rather than a satellite-to-phone side project.

The agreement, announced this week, covers Qualcomm's Dragonwing Cellular Infrastructure portfolio. Starlink Mobile will deploy a combination of radio equipment, baseband units and software to support cellular base station rollout at scale — specifically Massive MIMO radio units carrying what Qualcomm describes as advanced antenna capabilities, with claimed improvements in coverage, capacity and spectral efficiency.

The commercial commitment is the hard number here: "tens of thousands of sites deployed over the multi-year agreement." For context, that is the order of magnitude of site counts held by major US national operators, and Qualcomm is not shy about the comparison.

How big is the build, really?

Durga Malladi, EVP and GM for Technology Planning, Edge Solutions and Data Centre at Qualcomm, told Bloomberg the number of planned installations by Starlink Mobile will rival those deployed by major operators. On SpaceX's seriousness, he added: the company is "very committed and serious. Our degree of confidence is extremely high."

Those are confident words, and readers should weigh them accordingly. A multi-year agreement for tens of thousands of sites is a vendor contract, not an operating network; site acquisition, backhaul and handsets-in-the-field performance remain unproven until Starlink Mobile actually switches on. Neither company disclosed financial terms, deployment phasing beyond the 2027 target, nor how much spectrum Starlink Mobile will hold.

Notably absent from the announcement: any subscriber target, coverage percentage or throughput figure. What SpaceX has put on record is a date — nationwide deployment in 2027 — and a vendor contract sized like a tier-one operator's.

What does the Qualcomm technology actually cover?

The near-term deployment is terrestrial 5G. Qualcomm says the radio units, baseband and software in the deal support that generation today. The 6G language sits on top of the contract rather than inside it: Malladi said in a statement the two companies will "lay the foundations for AI-native 6G in the years ahead."

Qualcomm frames the energy-efficient processing and open, evolvable architecture as establishing "a foundation for future connectivity and intelligence." That is vendor positioning, and it should be read as such — the deployable product in this deal is 5G infrastructure. The 6G collaboration is a forward commitment with no attached timeline, spectrum plan or standard. With 6G standardization still in its early stages, that portion of the announcement is a statement of intent, not a deliverable.

Michael Nicolls, SVP of Starlink, positioned the deal as helping build "a fully-integrated nationwide satellite and terrestrial network that will keep people connected no matter where they are." That phrasing confirms the architecture: satellite direct-to-device service supplemented by conventional ground infrastructure, rather than a replacement for it.

Why does a satellite operator need cell towers?

Luke Pearce, principal analyst at CCS Insight, said the deal matches what the firm has predicted: "a conventional terrestrial network will need to complement satellite direct-to-device (D2D) services."

He cautioned that the project is a huge undertaking and the end product "will need to be competitive enough to justify the scale of investment" — a reminder that a tier-one-sized build carries tier-one-sized capital costs, and Starlink Mobile will have to win customers against incumbents to pay for it.

Pearce also sketched the international stakes: "If it works in the US, this hybrid model could scale internationally for Starlink and potentially be adopted by other satellite operators or smaller MNOs. It also creates significant opportunities for accelerating the uptake of open RAN and 6G, benefitting Qualcomm."

That last clause is worth noting for what it says about Qualcomm's motivation: an open, software-driven architecture across a greenfield national build is a showcase the chipmaker can point to as other operators weigh open RAN procurement.

The deal is the latest step in SpaceX's broader challenge to traditional US operators, following reported talks to acquire Boost Mobile network assets. Whether a satellite-terrestrial hybrid can compete on price and coverage against entrenched carriers will become testable from 2027, when Starlink Mobile says its nationwide service will be live.

Also reported

Share this article:

« PreviousNext »

More from Rebecca Stone

Rebecca Stone

Show full bio

Correspondent covering technology and product news in Carrier business, 5G and 6G, fiber and broadband at Telecom Gazette.

440 articles