The Synergies Between LEO Satellite Constellations and Submarine Cables

I’ve been exploring the impact of LEO satellite constellations on the wider telecom industry. Depending on the sector, the rise of LEO satellites is either complementary or competitive. Sometimes, the synergies are not evident. At other times, the competitive nature is masked by nascent requirements and emerging applications that could be game changers in the future. Underlying this, is the uncertainty of the LEO satellite constellation cost structure and commercial viability. In a recent paper we co-authored with Angola Cables, we analyzed the synergies between LEO satellites and submarine cable. Our approach was to look at the performance parameters of each service in the context of the cost structure. This provided the insights into potential LEO-Submarine synergies and competitive services which I summarize here.

The Submarine Cable Landscape

Submarine cables are the Internet’s backbone, carrying around 98% of the world’s traffic. Cycles of capacity gluts from overbuilding followed by retrenchments and M&As are highlights of the sector. The entry of cloud players and hyperscalars, such as Google and Facebook, over the past decade has drastically changed the market landscape. The consequent impact on both business models and technology, motivated submarine cable operators to seek new sources of revenue. For the sake of example, submarine cables connecting Africa operate at around 50% of capacity. Africa as a continent is well connected to the rest of the world with many subsea cables. The challenge lies in limited access networks that connect the interior of the continent, despite many fibre projects over recent years to connect populations hubs.

The LEO Landscape

I wrote in an earlier post about the differentiation among LEO constellations. Some constellations, such as SpaceX Starlink, target the consumer access play. Others, such as Telesat, target an enterprise play where requirements align with MEF services. Large spectrum allocations in Ku and Ka bands along with beamforming antenna array technologies allow a single LEO satellite to offer tens of gigabits of capacity.

	Submarine Cables	LEO Satellites
Capacity	10s – 100s Tbps	10s – 100s Gbps
Latency	~50 msec	~40 msec
Lifespan	25 years	5-10 years
Investment	$100-$500 Million	$5-$10+ Billion

The Cost Differential

We warp the performance comparison in the context of cost. The cost structure of LEO constellations is at least 10x higher than that of submarine cables for 10x less capacity (aggregate for the constellation). Submarine cables last 3-5x longer than satellites. With a cost differential at ~1000x per Gbps-month, the advantage of LEO satellites will centre on the highest value traffic.

With costs running into the billions of dollars, the success of LEO satellite constellations hinges on their ability to utilize the offered capacity to the fullest extent. Ironically, this is reminiscent of the conditions faced by submarine cable operators. Generating revenues is a common challenge shared by both sectors.

The Synergies

While satellite capacity pales with the terabit capacity of submarine cables, satellites could act as ‘feeder nodes’ into submarine cables. Thus, submarine cable operators could extend their services inland as opposed to terminating them at the landing station. Satellite operators would have a partner to close connectivity. Both get to fill their pipes, with maximal value attained when both submarine and satellite networks are integrated to preserve the end users’ traffic requirements.

Our paper outlines four synergetic use cases between satellites and submarine cables [here]. In summary, these use cases are:

1. Marke extension
2. Submarine cable offload
3. Submarine cable redundancy
4. Stop-gap solution

I won’t expand on these used cases here, because you could read the details in our paper.

Limited Competitive Grounds

There are a few cases where satellites can steal capacity from submarine cables. However, the traffic loss is minimal. It is possible for satellites equipped with Inter-satellite Links (ISL) to provide lower latency than submarine cables. This is attractive for applications such as high-frequency trading (HFT). However, HFT is important on a few routes. The mechanics of propagation and satellite constellation design limits the latency advantage to the longest of routes.

Concluding Thoughts

Satellite operators have played on the fringes of Internet. Satellite services focus providing connectivity services where terrestrial networks cannot reach, such as remote rural areas, maritime and aviation markets. Such applications would not directly be relevant to the submarine operator, and as such are not catalyst for closer integration. However, new markets are emerging from requirements for security and privacy. These applications could contribute to the satellite traffic load. It is an open question whether such applications could drive further synergies between satellites and subsea cables.

Land-space network convergence was the theme of the second Space Intersects Internet workshop we held on December 15, 2020. To download a summary transcript, enter your name and email in the boxes below.