Private 5G is reshaping how enterprises connect machines, people, and data at the edge – but success depends on bridging the long-standing divide between IT and OT. Globalstar’s XCOM RAN brings a software-defined approach to that challenge, combining open RAN architecture with licensed spectrum and satellite assets.
In sum – what to know:
Software and spectrum – private 5G platform from XCOM RAN blends licensed Band 53 with a software-defined RAN architecture for deterministic performance.
Reliability and scalability – the firm is seeking to bridge IT and OT silos and systems through dedicated on-prem connectivity that can be easily managed and scaled.
Physical AI and IoT – the logic is private 5G becomes the connective fabric for autonomous systems, real-time computer vision, and intelligent edge analytics
If Industry 4.0 – as a broad term to describe tech-geared digital change in industrial sectors – is to work, then the line between IT and OT has to blur, neatly and effectively; old data silos have to be smashed, and distributed hybrid architectures have to be engaged. With private 5G, some of this work is being done, necessarily, as IT and OT collaborate more closely to scope workloads and applications, and integrate new specialist operational (OT) connectivity into general corporate (IT) frameworks.
XCOM RAN, part of US satellite network operator Globalstar, has a twist on the standard private 5G fare, it reckons, insofar as it transposes parts of the radio access network (RAN) intelligence and control to the software, and packages it up as an easily programmable and scalable solution to make mission-critical Industry 4.0 smarter, faster. “We believe [it] is a superior private 5G solution,” says John Sweeney, vice president and head of global sales at the firm, speaking at Industrial Wireless Forum last week.
Parent company Globalstar, offering satellite comms in 120 countries, brings complementary assets into the bargain, as well. The firm has been operating a low-Earth orbit LEO constellation since the late 1990s. Sweeney recounts some of the XCOM RAN creation story – about its parent’s acquisition of mobile-satellite-service (MSS) licences in the S-band, including an 11.5 MHz chunk at 2.4 GHz (2483.5-2495 MHz), designed in 3GPP (for LTE/5G) as Band 53 (or n53 for 5G-NR) for Time-Division-Duplex (TDD) operations.
In 2023, Globalstar entered into an exclusive licensing deal with edge 5G startup XCOM Labs, led by former Qualcomm execs Paul Jacobs, Derek Aberle, and Matt Grob. Sweeney picks up: “They brought to the table revolutionary and innovative wireless technology for a next generation private 5G solution. So what we ended up with is a very powerful and unique combination… The XCOM RAN solution… is a spectrum agnostic solution, but comes with access to dedicated and licensed spectrum… backed with satellite assets.”
The pitch is that its satellite / terrestrial Band 5G spectrum combines with its private software-defined network (SDN) system, based on open RAN architecture, to create a differentiated proposition in the Industry 4.0 market. Sweeney says: “We chose open RAN because we wanted to make this software-defined. We wanted to do the software / hardware disaggregation and put the majority of the intelligence in the software.” By abstracting intelligence from the hardware layer, its private 5G solution gains agility and resilience, he argues.
“We took the intelligence – including some from the radios themselves – and put it in the distributed unit (DU). We use a coordinated multi-point radio system which centralizes the scheduling across different radios, which dramatically improves uplink and downlink performance – even in harsh or dense RF environments.” This coordinated architecture enables spatial massive MIMO with “advanced” beamforming, he says, delivering “higher spectral efficiency, extended coverage, higher availability, and improved mobility handoff”.
The “so-what” of all this, as Sweeney puts it, is about real industrial transformation. “It is about empowering industrial automation and robotics,” he says, citing the case of autonomous drones in various out-of-reach industrial climes – in “ports, inspecting hazardous cargo, you name it”, he says. In industrial environments, reliable (predictable, deterministic) performance on the uplink is as important as on downlink – and more important in most scenarios. “Something like 90 percent of data that’s collected in IoT sensors goes unused.”
He goes on: “It’s this data that [will] allow for intelligent edge analysis and closed loop automation… We’re focused on mission-critical operations… We’re not talking about [connectivity for] downloading TV shows in the airport lounge. This is about ensuring there is no downtime. Because if a production line or a distribution center shuts down or boats line up at a port because the communications are down, [then] you’re costing the enterprise tens of thousands if not hundreds of thousands of dollars in lost productivity per hour.”
Let’s break it down: the implications of private SDN-based systems in licensed spectrum in terms of flexibility and scale, IT/OT convergence, and future AI workloads. “[We] replace your rigid, static hardware-centric cellular system with a software-defined system where the radio, the core, and the edge functions all are [in] software… It lets enterprises and operators evolve and scale [the network] – just like any other IT application; pushing new capabilities, security updates, and performance tuning without actually having to touch the hardware itself.”
He adds: “It turns the RAN into something that is agile, programmable, and living – and constantly evolving and improving – rather than a fixed asset that is someday going to be forklift-upgraded.” The software shift effectively bridges the IT/OT divide, by making a private OT system compatible with general IT functions in the back office. “We’ve lived in a world of stovepipes for years. Software-defined private 5G creates a single fabric that connects all the systems, machines, devices, sensors, and people with deterministic performance and low latency.”
Because it is architecturally-private, still – as prescribed in the design, as deployed on the shop floor – the network’s original OT remit is satisfied. “All that [IoT] telemetry data stays on premise; which means OT maintains ownership, security, data sovereignty – all of which is critical. And then IT gains a standards based system that is perfect for automation, analytics, and AI.” Sweeney describes it as a new IT/OT “convergence layer” that “OT trusts for reliability and safety” and “IT trusts for integration, security, and scalability”.
He says: “Because the baseband sits on a COTS (commercial off-the-shelf) server, scaling the network is really like scaling a data center application. All you do is add compute, deploy coverage, or push features into the applications – just like anything else via software.” The model “eliminates proprietary hardware lock-in, and makes the lifecycle management into more of a continuous development, and continuous delivery engine,” he says. The result is both “a lower cost of ownership” and the ability to “innovate at the speed of software”.
Plus, by virtue of Globalstar’s Band 53 holdings (also let to third-party private 5G vendors), all the deterministic 5G goodness, made effective in an IT-friendly and OT-optimised network system, is reinforced in licensed spectrum. Spectrum remains a critical differentiator, he says. “We are spectrum agnostic; we can do CBRS, we can do pretty much any TDD spectrum – and we prefer TDD because it helps with the uplink and the downlink flexibility…. [But] CBRS is shared, and Wi-Fi is unlicensed, so [performance depends on] who else is swimming in the same lane.”
Enterprises have their “own private lanes”, he says. “Band 53 gives enterprises or operators licensed interference-free spectrum… Performance is completely predictable, service is assured – even in the noisiest RF environments.” The XCOM RAN system is proven, he suggests. Real-world tests and deployments have shown the biggest challenges aren’t always technical. “Because the technology works, the surprise isn’t so much for us, but for the users – and it really is cultural,” says Sweeney.
He explains: “When OT and IT, and all the automation vendors, get on the same page, on the same private 5G network, then they start solving problems together that were impossible to do with Wi-Fi or siloed systems. And the moment they see deterministic wireless replacing cables, [and to deploy] autonomous forklifts, real-time computer vision, and data analytics at the edge, then you see the cultural shift to embrace, adopt, accelerate IoT and physical AI at a much more rapid pace.” And so where does this end up?
Looking ahead, Sweeney foresees “multiple [private 5G] super cells” covering millions of square feet, supporting “AI orchestrations”, supported by “satellite backhaul” – all available to enterprises as a simple software upgrade rather than a complex labour intensive “forklift rebuild” of hardware-heavy network systems. He says: “The future for autonomous networks is the ability to self-tune for latency, mobility, energy, and interference. We’re no longer talking about five robots or 10 robots at a plant; we’re now talking about hundreds if not thousands of them.”
The global economy is entering a new AI era, he suggests. “We’re looking at an inflection point that will surpass the internet [revolution] at the turn of the century. It is a seismic shift, not just in how factories and ports operate, but how all of society lives, works, and plays.” But this AI revolution needs to be connected in private workplaces and public venues of all kinds – in a way that is reliable and trusted. Sweeney says: “It all falls apart if you don’t have reliable and secure comms to tie it all together, and to keep up with the speed of innovation.”
