Two Engineering Engagements Target D2D Satellite Connectivity
Aug 24, 2026
AccelerComm has wrapped up two separate engineering engagements with Eclipse Space, the sovereign satellite infrastructure company that emerged from stealth mode earlier this year. The work focused specifically on defining the physical-layer architecture for Eclipse’s next-generation spacecraft, covering both its direct-to-device and broadband payload lines. AccelerComm announced the completed engagements, framing the collaboration as an early but important step in Eclipse’s broader plan to deliver turnkey, customer-owned satellite constellations to governments and enterprises that want more control over their own communications infrastructure.
Credit: AccelerComm
The engineering work centered on Eclipse’s next-generation digital beamformer, a component designed to handle both direct-to-device and broadband connectivity from a single scalable platform rather than requiring separate hardware for each. That shared architecture matters for Eclipse’s fabless production model, since keeping one common spacecraft design across multiple service types simplifies manufacturing even as the company works with outside contractors to actually build the satellites. AccelerComm contributed its 5G non-terrestrial network Layer 1 physical-layer IP to the project, including error correction and waveform optimization technology that will now underpin Eclipse’s modem architecture going forward, with the resulting system built to support standards-aligned Narrowband NTN and 5G New Radio NTN services.
Physical-layer work on direct-to-device connectivity is brutal in a way that’s easy to underestimate from the outside. A signal has to cross hundreds of kilometers between a satellite and a phone with an antenna the size of a thumbnail and barely any transmit power to work with, so beamforming, signal processing, and error correction all have to work near flawlessly just to keep a connection alive without draining a spacecraft’s limited power budget. AccelerComm CEO David Helfgott framed it simply. His company’s expertise sits precisely at that layer, and flight-proven technology already running in orbit gave Eclipse a reason to bring in outside help rather than solve the problem from scratch on its own timeline.
Eclipse has been under pressure to nail down its hardest engineering calls before its demonstration mission, and that pressure shaped how this partnership came together. Get the physical-layer architecture wrong now, and Eclipse risks an expensive redesign later, so handing that piece to a specialist frees up the company’s own engineers for the systems integration work that a fabless, customer-owned constellation model demands. Derek Huerta, Eclipse’s co-founder and CEO, put it in blunt, practical terms, saying the collaboration sharpened his team’s payload specifications and gave them more confidence heading into both the demonstrator and whatever operational constellations follow. That’s a telling way to describe it for a company whose whole pitch depends on moving fast without owning a single factory.
There’s also a business signal buried in this deal that goes beyond the technical details. AccelerComm built its name supplying physical-layer IP into established 5G NTN and satellite deployments, the kind of work that usually means selling into one big centralized network at a time. Eclipse is a different animal entirely, building sovereign, customer-owned infrastructure meant to be replicated across many separate government and enterprise buyers rather than run as one unified system the way SpaceX or a traditional GEO operator would do it. Both sides say they expect to keep working together as Eclipse pushes toward its demonstration mission and eventually scales into full operational deployment, which would put AccelerComm squarely inside a market that barely existed a few years ago, distributed constellations owned and controlled by the countries and companies that actually use them.
