How does satellite networking fit into the 5G roadmap?

Indian operator Reliance Jio Infocomm last week broadened debate about satellite services when it announced a joint venture with SES to provide connectivity at what it describes as “highly affordable prices”.

Branded as JioSpaceFiber, the service will access SES’s medium earth orbit (MEO) satellite technology, to provide what it claims is India’s first satellite-based gigabit service.

Jio said this new service will be available nationwide to serve rural regions in addition to its existing broadband services, to help bridge the digital divide. JioSpaceFiber will also provide additional capacity for 5G mobile backhaul, a more traditional use of satellite in areas where fiber or microwave backhaul is not available or cost-effective.

Typically, satellite connectivity has been considered an expensive alternative to terrestrial services, suitable only in extreme situations beyond the geographic reach or cost feasibility of regular mobile or fixed networks. But two factors are combining to push communications service providers (CSPs) to review their satellite-based connectivity strategies:

• Non-Terrestrial Networks (NTNs), based on Low Earth Orbit (LEO) satellite systems, have become much easier for CSPs to utilize, due to a direct terminal-to-satellite air interface defined in 3GPP standard Release 17, integrating satellite networks and terrestrial 5G networks, to support ubiquitous coverage.
• The large amount of LEO infrastructure investment – not to mention industry media headlines – in the past five years. LEO satellites orbit much closer to the earth’s surface – typically under 1,500 kilometers – compared to Geostationary Earth Orbit (GEO) satellites, which orbit around 22,000 kilometers. The closer distance means broadband connectivity from LEO satellites has significantly less latency; consequently, LEO is on track to become the dominant technology option for telecommunications, with NASA predicting as many as 50,000 LEO satellites in orbit in ten years’ time.

Investment has been driven by companies such as Elon Musk’s SpaceX (which provides broadband internet globally through its Starlink service), AST SpaceMobile, Lynk, OneWeb, Sateliot and TeleSat. Amazon has also joined in the race to launch its own satellites into low Earth orbit, via its Project Kuiper, beginning in 2024.

The European Union is getting in on the act too, announcing its intention to launch its own satellite internet constellation, Infrastructure for Resilience, Interconnection and Security by Satellites (IRISS), with the “ambition” for initial services in 2024 and full operational capacity in 2027. Intended to be multi-orbital connectivity infrastructure, it will include the launch of up to 170 satellites into low Earth orbit between 2025 and 2027. This will provide what the EU describes as “sovereign, secure, resilient and cost-effective seamless communication services”.

The constellation is intended to provide broadband connectivity to the whole of Europe, including current connectivity dead zones, as well as to the whole of Africa. It will also build on developing capabilities such as quantum encryption satellites and new disruptive technologies.

New commercial models

The accumulation of satellite constellations designed to democratize the availability and cost of connectivity raises the question of how CSPs can use NTNs within their networks and services and how new commercial models will impact the competitive landscape.

Integrated mobile-plus-satellite networks that enable on-demand ubiquitous access could support service scenarios such as “NR-NTN”, using the 5G New Radio (NR) framework to enable smartphones to directly connect to satellites to provide data voice services; and “IoT-NTN”, offering satellite IoT services based on low-complexity, enhanced machine-type communications (eMTC) using narrowband IoT (NB-IoT) terminals. From a user perspective this means smartphones or other devices could connect directly to satellites from previously unconnected locations, providing ubiquitous coverage to underserved users or supporting new IoT applications. Additionally, NTNs can provide a layer of resilience and redundancy to the existing 5G network, such as providing backup connectivity to ensure continued service for mission-critical communications in the event of natural disasters, regional conflicts or network outages.

Regarding IoT-NTN, Barcelona-based constellation network operator Sateliot has launched the first of 250 small (10kg) satellites designed to communicate with terrestrial cell sites to deliver access to Internet of Things (IoT) services. Sateliot envisions applications for multiple public and private markets including road, rail, air and sea transportation.

Other possibilities range from rural business applications such as agriculture, forestry and mining to automotive apps, industrial IoT services, logistics and asset tracking and remote healthcare. Sateliot hopes to enable users to seamlessly switch between terrestrial networks to a non-terrestrial 5G network, without any need to purchase any extra hardware, such as antennas or modems, using current SIMs and negotiating standard roaming agreements. As its first satellite begins operation, Sateliot is already boasting sales of more than $1.3 billion.

Straight to your smartphone?

Potentially the most disruptive manifestation of satellite in terrestrial service markets is the advent of direct-to-mobile smartphone services, in which a LEO satellite constellation connects directly to mobile devices using existing terrestrial spectrum and equipment. These are now starting to emerge. For example, US-based AST SpaceMobile is currently building a satellite network capable of voice and data communications with standard 4G/5G smartphones and other cellular devices, backed by telco partners and investors including AT&T, American Tower, Rakuten Mobile and Vodafone. The company says it has signed agreements and documents of understanding with more than 35 mobile network operators to provide them with wholesale services, including Orange, Globe and MTN.

AST SpaceMobile claims it is developing “the first space-based cellular broadband network” accessible directly by standard mobile phones without the need for any modifications or to purchase or install additional equipment. This contrasts with the “traditional” satellite approach, such as the Starlink model, in which signals are beamed from space to receivers on the ground, serving as a hub for people to connect other devices to – with dishes costing approximately $600 (despite including a subsidy), limiting its potential market.

But the direct-to-mobile approach raises power requirement questions because staying connected on a standard satellite-connected smartphone brings battery performance challenges, including the possibility of the device overheating. AST SpaceMobile has stated that because its satellites are designed to function like ordinary cell sites, power usage by mobile devices is akin to that of a phone in a rural area connecting to a typical macro cell tower.

There are other challenges too, such as the need to balance coverage and performance. The network's performance, in terms of latency and capacity, is affected by distance and path loss. Beaming signals to extremely wide areas can only be achieved with low-frequency spectrum, often below 1 GHz, and the long distance through the atmosphere that the signal must travel between device and satellite often leads to higher latency compared with terrestrial alternatives.

A combination of these factors and a greater susceptibility to interference has always meant that satellite networks face a range of performance challenges, hence the limited uptake seen up until now. LEO constellation builders claim the latency issue is not as significant in LEO systems compared to geostationary satellites, but it remains unclear whether such latency would be perceptible to users. Ultimately, many of these challenges come down to the antenna technology, with a commercial 4G or 5G antenna featuring 64 elements. However, new advanced intelligent antenna solutions, including innovations such as multibeam and lighter, more scalable digitally phased array antennas, can simultaneously transmit hundreds of beams, narrowing these so finely and accurately that it creates the equivalent of a personal cell site for the user. This enables NTNs to be economically viable, and able to integrate seamlessly into terrestrial 5G networks.

Going it alone?

Alongside the technical evolution driving better performance, and rapidly accelerating satellite availability, an evolution of potential commercial service models is also underway. For satellite providers, working with operators on a wholesale basis rather than going to customers directly seems most logical, as it avoids the need to buy spectrum and allows them to access the operator’s existing customer relationships. This arrangement also makes sense for operators seeking to cost-effectively connect customers living in rural and remote areas.

To date the commercial landscape is being shaped by partnerships between terrestrial and satellite CSPs, but could NTNs bypass terrestrial mobile and broadband networks altogether and offer competing services?

Most satellite providers are at pains to emphasize that NTNs are not an alternative to terrestrial cellular or fixed-line connectivity – they are a complement, either as a wholesale play or filling in coverage gaps where those networks have no points of presence. Musk, however, has stated that Starlink’s long-term goal of providing gigabit-level, low latency connectivity could be a potential competitor to consumer broadband and mobile services in the future.

Though this sort of competitive stance from satellite providers seems unlikely where they have wholesale customers, there remains the potential for market disruption in both home broadband and consumer mobile segments. After all, even though they may not be able to rival terrestrial networks for performance in terms of capacity for broadband, satellite-based services could represent an attractive alternative in developing markets – not “superfast” connections necessarily, but “fast enough” if a market has previously been underserved by existing broadband options.

If successful, satellite broadband service providers may feel inclined to extend their direct-to-end user presence into other locations where they do have existing competitors. Their cost base, and need for recurring revenue, may mean they have to.